System

Abstract

A system is provided.SOLUTION: A system comprising: means for inputting and transmitting a question from a terminal by a user; means for receiving the question and analyzing the question by natural language processing by a server; means for retrieving an employee rule and a past case from a database based on an analysis result; means for generating an answer based on a retrieval result and shaping the answer by natural language generation; and means for transmitting the generated answer to the terminal and displaying the 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] Currently, it takes a lot of time and effort for employees to check employee regulations. Specifically, they have to find the relevant clause in the work regulations and are left unsure about how to interpret it. In addition, when employees inquire with human resources personnel, they are unable to obtain accurate information immediately, which reduces work efficiency. To solve these issues, a system that can quickly and accurately provide employee regulations and past cases is needed. [Means for solving the problem]

[0005] To solve the above problems, the present invention provides a system that includes a means for a user to input and send a question from a terminal, a means for a server to receive the question and analyze the question using natural language processing, a means for searching a database for employee regulations and past cases based on the analysis results, a means for generating an answer based on the search results and formatting it using natural language generation, and a means for transmitting the generated answer to the terminal and displaying it to the user. This system allows employees to receive quick and accurate answers to their questions, thereby deepening their understanding of the regulations and improving work efficiency.

[0006] "User" refers to an employee who uses the system to ask questions or make inquiries.

[0007] "Terminal" means a computing device through which a user accesses the system and displays questions and answers, including, but not limited to, a PC, smartphone, or tablet.

[0008] A "question" refers to a text input that a user makes to a system to confirm information or request instructions.

[0009] A "server" refers to a computer system that receives data sent from a terminal, processes it, and provides the necessary information to the user.

[0010] "Natural language processing" refers to the technology in which a server analyzes questions received from users and extracts keywords and intent.

[0011] "Employee regulations database" refers to data storage that stores regulations and rules for employees.

[0012] A "past case database" refers to data storage that records past cases and how they were handled.

[0013] "Analysis results" refers to keywords and intent extracted through natural language processing.

[0014] "Search" refers to the process of retrieving relevant information from the employee regulations database and past case database based on the analysis results.

[0015] "Answer" refers to information that the server generates based on the search results and provides to the user.

[0016] "Natural language generation" refers to the technology that formats the answers generated by the server into a format that is easy for the user to understand.

[0017] "Display" refers to the process by which the terminal visually shows the answer to the user's question. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0026] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0039] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases. The detailed implementation method for users to input questions and receive appropriate answers is described below.

[0040] First, the user accesses the system using a device such as a PC or smartphone and enters their question. For example, they might enter, "Please tell me how to register attendance when I go to the hospital." This question is immediately sent from the device to the server.

[0041] The server uses a natural language processing (NLP) module to analyze the questions received from the device. The server analyzes the questions and extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. This makes it clear which regulations should be referenced.

[0042] The server then searches the employee regulations database and past case database for related information based on the extracted keywords and intent, and retrieves the relevant employee regulations and similar past cases.

[0043] The server evaluates the search results and generates the most appropriate answer. The generated answer is formatted using a natural language generation (NLG) module to make it easy for the user to understand. For example, the server might generate an answer such as, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0044] Finally, the server sends the formatted answer to the terminal, and the terminal displays the generated answer to the user, allowing the user to quickly obtain appropriate information for the question, deepening their understanding of employee regulations and improving work efficiency.

[0045] As a concrete example, consider the case where a user asks, "What is the procedure for taking paid leave?" In this case, the server extracts keywords such as "paid leave," "taking," and "procedure," and searches for relevant regulations in the employee regulations database. It then takes into account similar past cases and generates an answer showing the appropriate procedure. For example, an answer including specific steps, such as "To take paid leave, please obtain approval from your supervisor and then apply through the attendance management system," is generated and presented to the user via their device.

[0046] As described above, the system of the present invention allows employees to get quick and accurate answers to their questions. This system significantly reduces the time and effort required to check employee regulations, thereby improving work efficiency.

[0047] The processing flow will be explained below.

[0048] Step 1:

[0049] User: Accesses the system and inputs a question into the interface on the terminal. For example, inputs "How do I register my attendance when I go to the hospital?"

[0050] Terminal: Receives user input and sends questions to the server.

[0051] Step 2:

[0052] Server: Receives questions sent as HTTP requests from the terminal.

[0053] Server: Passes the received question to a natural language processing (NLP) module to analyze the text.

[0054] Specific operation: Divide the question into tokens and extract keywords (e.g., "visiting the hospital" or "attendance registration") and intent.

[0055] Step 3:

[0056] Server: Based on the analysis results, a query is generated to search the employee regulations database and the past case database.

[0057] Specific operation: Generates SQL queries, etc., and retrieves information related to the question from the database.

[0058] Step 4:

[0059] Server: Aggregates information retrieved from the database and evaluates search results.

[0060] Specific operation: The system sorts the data and selects the most relevant information, such as "A doctor's note is required when visiting the hospital."

[0061] Step 5:

[0062] Server: Generates an answer to the user's question based on the evaluation results.

[0063] How it works: Responses are constructed based on employee policies and past experience, and formatted in a way that is easy for humans to understand. A natural language generation (NLG) module assists in this process.

[0064] Example: "When visiting a hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0065] Step 6:

[0066] Server: Sends the generated answer to the device.

[0067] Specific operation: The formatted response is returned to the terminal in JSON format or similar.

[0068] Step 7:

[0069] Terminal: Displays the answer received from the server on a user interface.

[0070] Specific behavior: Updates the UI to present the received information to the user in the appropriate format.

[0071] Step 8:

[0072] User: Check the answers displayed on the device and take action if necessary.

[0073] Example: Take specific action such as registering as "visiting the hospital" in the attendance management system.

[0074] Through these steps, users can get instant and accurate answers to their questions, streamlining the employee policy review process and improving work productivity.

[0075] Example 1

[0076] 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."

[0077] In modern companies, having employees be able to quickly and accurately refer to employee regulations and past cases directly leads to improved work efficiency. However, current systems require employees to manually search for regulations and cases, which is time-consuming and labor-intensive. To solve this problem, there is a need for a system that allows users to obtain appropriate answers simply by entering a question.

[0078] 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.

[0079] In this invention, the server includes a means for a user to input and send a query from a computer terminal, a means for the server to receive the query and analyze it using natural language processing, and a means for searching for regulations and past cases from a database based on the analysis results, thereby enabling a user to quickly and accurately refer to employee regulations and past cases.

[0080] A "user" is a person who operates a computer terminal to input a query and obtain a response from the system.

[0081] "Computer terminal" refers to a device that can connect to the Internet, such as a PC or smartphone.

[0082] A "query" is a question or command that a user enters into a system with the purpose of obtaining information.

[0083] A "server" is a computer system that receives inquiries, performs natural language processing and search processing, and generates and transmits answers.

[0084] "Natural language processing" is a technique used by computers to analyze human language and understand user queries.

[0085] "Analysis" is the process by which the server understands the content of the query and extracts relevant information.

[0086] A "database" is a collection of information in which information such as regulations and past cases is structured and stored.

[0087] "Policies" are documents or information that describe company rules and procedures that employees must follow.

[0088] "Past cases" are records of responses and actions taken in response to similar inquiries that have occurred in the past.

[0089] "Searching" is the process of finding specific information within a database.

[0090] A "response" is information or instructions that the server generates and sends in response to a user's inquiry.

[0091] "Generation" is the process by which the server creates a response for the user based on the analysis results.

[0092] "Natural language generation" is a technology that allows a computer to create text in a form that humans can understand.

[0093] "Transmission" means the act of electronically transferring a query or response from one party to another.

[0094] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases. Specific embodiments are described below.

[0095] Hardware and software used

[0096] This system mainly consists of the following hardware and software:

[0097] Hardware

[0098] 1. Computer terminal

[0099] A PC or smartphone for users to enter questions

[0100] 2. Server

[0101] A high-performance computer for analyzing questions, searching databases, and generating answers

[0102] software

[0103] 1. Natural Language Processing (NLP) Module

[0104] Software used to analyze user questions. Examples include NLP libraries such as spaCy and Transformers.

[0105] 2. Database Management System (DBMS)

[0106] It is used to execute SQL queries to search for employee policies and past cases. Examples include MySQL and PostgreSQL.

[0107] 3. Natural Language Generation (NLG) Module

[0108] It is used to generate answers based on search results and format them in a format that is easy for users to understand. Specific examples include OpenAI's GPT-3 and GPT-4.

[0109] Data processing and data calculation

[0110] 1. Question analysis

[0111] The question entered by the user on the computer terminal is sent to the server, which uses an NLP module to extract keywords such as "hospital visit" and "attendance registration" from the question and the user's intention.

[0112] 2. Database search

[0113] The server searches for relevant information from the employee regulations database and past case database based on the extracted keywords. This search uses SQL queries. For example, the SQL query for the keyword "hospital visit" is executed as follows: "SELECT FROM employee_regulations WHERE keyword='hospital visit'".

[0114] 3. Answer generation

[0115] The server evaluates the search results and generates the most appropriate answer using an NLG module, which is formatted in a way that is easy for the user to understand, such as "To take paid leave, please get your supervisor's approval and apply through the attendance management system."

[0116] Specific examples

[0117] A specific example will be given below in which a user asks on a computer terminal, "Please tell me how to register attendance when visiting a hospital."

[0118] 1. Enter your question

[0119] Users access the system using a PC or smartphone and enter their questions.

[0120] 2. Submit a question

[0121] The terminal immediately sends the query to the server.

[0122] 3. Question analysis

[0123] The server uses an NLP module to analyze the question and extract the keywords "hospital visit" and "attendance registration."

[0124] 4. Database Search

[0125] The server searches the employee regulations database and past case database for relevant information and acquires the necessary information.

[0126] 5. Answer generation

[0127] Based on the search results, the server uses the NLG module to generate a response such as, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0128] 6. Submit your response

[0129] The server sends the generated answer to the computer terminal, which displays it to the user.

[0130] Prompt Sentence Examples

[0131] "Please tell me how to register attendance when visiting the hospital."

[0132] What is the procedure for taking paid leave?

[0133] "I would like to know how to claim expenses when working from home."

[0134] By entering the above prompts, the user can quickly and accurately obtain information on employee regulations and past cases. The system of the present invention provides prompt and appropriate answers to the user's questions, significantly improving the work efficiency of employees.

[0135] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0136] Step 1:

[0137] A user accesses the system using a computer terminal and inputs a question. Specifically, the user opens a browser, accesses the system's web interface, and inputs a question into the input form. For example, the user might input, "Please tell me how to register attendance when visiting the hospital." This question becomes input data.

[0138] Step 2:

[0139] The terminal immediately sends the question entered by the user to the server. Specifically, when the user clicks the send button, the terminal sends the input data to the server as an HTTP POST request. This transmission becomes the input for processing.

[0140] Step 3:

[0141] The server analyzes the questions received from the device using a natural language processing (NLP) module. Specifically, the server tokenizes the questions received using an NLP library (e.g., spaCy or Transformers) and performs semantic analysis. This analysis extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. The analysis results are the output for data processing.

[0142] Step 4:

[0143] The server searches for relevant information from the employee regulations database and past case database based on the analysis results. Specifically, it generates an SQL query based on the extracted keywords and searches for data using a database management system such as MySQL or PostgreSQL. For example, an SQL query such as "SELECT FROM employee_regulations WHERE keyword='hospital visit'" is executed. The data retrieved by this search becomes the output of data calculations based on the input data.

[0144] Step 5:

[0145] The server evaluates the search results and generates the most appropriate answer. Specifically, it uses an NLG module (for example, OpenAI's GPT-3) to generate natural language based on the search results, and creates an answer in a format that is easy for the user to understand. For example, the answer could be something like, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system." This generated answer becomes the output data.

[0146] Step 6:

[0147] The server sends the generated answer to the terminal. Specifically, it organizes the generated answer in JSON format and sends it to the terminal as an HTTP response. This transmission becomes the output of the processing.

[0148] Step 7:

[0149] The device displays the answer received from the server to the user. Specifically, it updates the DOM of the web page and reflects the received text in the answer display area. As a result, the user can see the appropriate information for the question on the screen. This display is the final output.

[0150] (Application example 1)

[0151] 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."

[0152] In brick-and-mortar stores, when store clerks or managers have questions about store operations, it can be difficult to quickly and accurately obtain the appropriate information. For example, when they need to immediately check product return policies or complaint handling procedures, not being able to obtain the appropriate information in a timely manner can reduce work efficiency. Another issue is the lack of a way to easily check information on a device at hand while working on other tasks.

[0153] 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.

[0154] In this invention, the server includes means for a user to input and send a question from a terminal, means for the server to receive the question and analyze the question using natural language processing, means for searching a database for regulations and past cases based on the analysis results, means for generating an answer based on the search results and formatting it using natural language generation, and means for transmitting the generated answer to the terminal and displaying it to the user, thereby enabling users to quickly and accurately obtain appropriate information for questions about store operation in a physical store.

[0155] "Means for users to input and send questions from a terminal" refers to the process and interface by which a user inputs a specific question using a smart device and sends the question to a server over a network.

[0156] "Means for the server to receive a question and analyze the question using natural language processing" refers to a module or algorithm that allows the server to receive a question sent by a user, analyze the question using natural language processing technology, and understand its meaning and intent.

[0157] "Means for searching for regulations and past cases from a database based on the analysis results" refers to functions and modules that extract keywords from the results of the analyzed questions and use them to search a database in which regulations and past cases are stored.

[0158] "Means of generating answers based on search results and formatting them using natural language generation" refers to the function of using natural language generation technology to create answers in a format that is easy for users to understand based on information obtained from a database, and then formatting them.

[0159] "Means for transmitting the generated answer to the terminal and displaying it to the user" refers to the processes and techniques for transmitting the formatted answer back over the network to the originating smart device and visually displaying it to the user.

[0160] "A means of providing work support by displaying generated answers on a smart device" refers to a function that displays generated answers on smart devices such as smartphones, tablets, and smart glasses, and supports work related to store operations.

[0161] "Means of utilizing regulations related to store operations and databases of past cases" refers to the ability to utilize databases that record regulations related to store operations (such as refund policies and complaint handling procedures) and specific past cases.

[0162] This invention provides a question-answering system that uses AI technology to improve the efficiency of brick-and-mortar store operations. Specific embodiments of this system are described below.

[0163] First, a user (a store clerk or manager) uses a smart device (smartphone, tablet, smart glasses, etc.) to input and send a question. For example, they might input, "Please tell me the procedure for returning a product." This question is immediately sent to the server via the network.

[0164] The server uses a natural language processing (NLP) module such as Spacy to analyze the received question. Specifically, it extracts keywords such as "return" and "procedure" from the question and understands the user's intent.

[0165] The server then searches a database for relevant regulations and past cases based on the extracted keywords. The regulation database includes product return policies and store operation rules, while the past case database stores specific past cases. This allows for quick retrieval of the relevant information.

[0166] The server then generates an answer based on the information it has obtained, using natural language generation (NLG) modules like Hugging Face Transformers to format the answer in a way that is easy for the user to understand, such as specific instructions like "When returning an item, please bring the item with the receipt and fill out the return form."

[0167] The generated answer is sent to the original smart device via the network, where the user can visually check it and take any necessary action immediately, significantly improving the efficiency of store operations.

[0168] For example:

[0169] When a user asks, "What is the procedure for returning a product?", the server extracts keywords such as "return" and "procedure" and searches the database for relevant regulations and past cases. It then generates a response such as, "When returning a product, please bring the product along with the receipt and fill out the necessary information on the return form," which is displayed on the smart device. This process allows the user to take prompt and appropriate action.

[0170] Example prompt sentence:

[0171] Q: What is the procedure for returning a product?

[0172] A: When returning an item, please bring the item with you along with your receipt and fill out the return form.

[0173] This system makes it possible to quickly resolve questions about store operations and improve operational efficiency.

[0174] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0175] Step 1:

[0176] A user uses a smart device (smartphone, tablet, smart glasses, etc.) to input a question and press the "Send" button. For example, the user might input "Please tell me the procedure for returning a product." At this time, the input question is sent as text data from the device to the server.

[0177] Input: The question text entered by the user (e.g., "How do I return an item?")

[0178] Output: Submitted question text

[0179] Step 2:

[0180] The server receives the question sent from the device and analyzes it using a natural language processing (NLP) module that utilizes Spacy. Specifically, the question is tokenized and keywords such as "return" and "procedure" are extracted. The user's intent is understood based on the extracted keywords.

[0181] Input: Submitted question text

[0182] Output: Extracted keywords and user intent (e.g., "return," "procedure")

[0183] Step 3:

[0184] The server searches the regulatory database and past case database based on the extracted keywords, thereby obtaining relevant regulatory information on store operations and past case studies that have previously addressed similar questions. The search results are stored in list format.

[0185] Input: Extracted keywords and user intent

[0186] Output: Search result list (e.g. related regulatory information and past cases)

[0187] Step 4:

[0188] Based on the search results, the server uses a natural language generation (NLG) module that utilizes Hugging Face Transformers to generate answers in a format that is easy for the user to understand, specifically outputting answers in natural-sounding sentences while taking into account the context of the search results.

[0189] Input: Search result list

[0190] Output: The generated answer (e.g., "When returning an item, please bring the item with you along with your receipt and fill out the return form.")

[0191] Step 5:

[0192] The server sends the generated answer to the original smart device via the network, and the smart device displays the received answer on its interface so that the user can visually confirm it.

[0193] Input: Generated Answer

[0194] Output: Answer displayed on the user's terminal

[0195] Step 6:

[0196] The user checks the answers displayed on the smart device and takes specific action, such as actually returning the product based on the answers.

[0197] Input: Answer displayed on smart device

[0198] Output: User's actual action (e.g., returning a product)

[0199] In this way, by responding quickly and accurately to questions about running a physical store, the operational efficiency of the entire store can be improved.

[0200] 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.

[0201] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases, and by combining it with an emotion engine that recognizes the user's emotions, it is possible to obtain more appropriate answers. Specific embodiments of the present invention are described below.

[0202] First, the user accesses the system using a device such as a PC or smartphone and enters a question. For example, they might enter, "Please tell me how to register attendance when I go to the hospital." This question is immediately sent from the device to the server.

[0203] The server first uses an emotion engine to analyze the question received from the terminal. The emotion engine analyzes emotions (e.g., stress, anxiety, anger, joy, etc.) from the user's input text and obtains the results. For example, if the emotion engine detects "anxiety" from the user's question, that information is used in the subsequent analysis process.

[0204] The server then passes the acquired emotional information along with the question to a natural language processing (NLP) module. The server analyzes the question and extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. This makes it clear which regulations should be referenced.

[0205] The server searches the employee regulations database and past case database for related information based on the extracted keywords and intent, and obtains the relevant employee regulations and similar past cases.

[0206] The server evaluates the search results and generates the most appropriate answer. The generated answer is adjusted to reflect the user's emotional information. For example, if the user is feeling anxious when asking a question, the answer will include phrases such as "Don't worry" or "Don't worry," providing an answer that takes the user's emotions into consideration. This is supported by a natural language generation (NLG) module.

[0207] Finally, the server sends the formatted answer to the terminal, and the terminal displays the generated answer to the user. This allows the user to quickly obtain appropriate information for the question and receive an answer that takes into consideration their feelings, thereby deepening their understanding of employee regulations and improving work efficiency.

[0208] As a concrete example, consider the case where a user asks, "What is the procedure for taking paid leave?" and the emotion engine detects "stress." The server extracts keywords such as "paid leave," "taking," and "procedure," and searches for relevant regulations in the employee regulations database. It then generates an answer showing the appropriate procedure, taking into account similar past cases. For example, an answer that takes specific steps and emotions into consideration, such as "To take paid leave, please get your supervisor's approval and then apply through the attendance management system. It's okay, your health is important, so please take plenty of rest," is generated and presented to the user via their device.

[0209] As described above, by recognizing the user's emotions and adjusting responses accordingly, the system of the present invention not only allows employees to get quick and accurate answers to their questions, but also allows for more user-friendly responses by taking emotions into consideration, thereby streamlining the confirmation process for employee regulations and improving work productivity.

[0210] The processing flow will be explained below.

[0211] Step 1:

[0212] User: Accesses the system using a device such as a PC or smartphone and enters a question. For example, "Please tell me how to register attendance when going to the hospital."

[0213] Terminal: Receives user input and sends questions to the server.

[0214] Step 2:

[0215] Server: Receives questions sent as HTTP requests from the terminal.

[0216] Server: Passes the received question to the emotion engine and analyzes the user's emotion from the text.

[0217] Specific operation: Analyzes the question and applies an emotion model to extract emotions such as "stress," "anxiety," and "joy." For example, the emotion "anxiety" can be detected from the question "Please tell me how to register attendance when going to the hospital."

[0218] Step 3:

[0219] Server: Holds the extracted sentiment information and passes the question to a natural language processing (NLP) module to analyze the text.

[0220] Specific operation: The question is divided into tokens, and keywords (e.g., "visiting the hospital" or "attendance registration") and the user's intent are extracted.

[0221] Step 4:

[0222] Server: Based on the analysis results, a query is generated to search the employee regulations database and the past case database.

[0223] Specific operation: Generates SQL queries, etc., and retrieves information related to the question from the database.

[0224] Step 5:

[0225] Server: Aggregates information retrieved from the database and evaluates search results.

[0226] Specific behavior: Orders the retrieved data and identifies the most relevant information, such as "A doctor's note is required when visiting the hospital."

[0227] Step 6:

[0228] Server: Based on the evaluation results, it generates an answer to the user's question. It adjusts the answer taking into account emotional information.

[0229] How it works: The server constructs a response based on employee policy and past case data, and a natural language generation (NLG) module formats it. For example, if anxiety is detected, a phrase like "Don't worry" is added to the response.

[0230] Example: "When you visit the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system. Don't worry, everything will be fine."

[0231] Step 7:

[0232] Server: Sends the generated answer to the device.

[0233] Specific operation: The formatted response is returned to the terminal in JSON format or similar.

[0234] Step 8:

[0235] Terminal: Displays the answer received from the server on a user interface.

[0236] Specific behavior: Updates the UI to present the received information to the user in the appropriate format.

[0237] Step 9:

[0238] User: Check the answers displayed on the device and take action if necessary.

[0239] Example: Take specific action such as registering as "visiting the hospital" in the attendance management system.

[0240] This allows users to receive instant, accurate answers to their questions, and also allows them to receive answers that take into consideration their feelings, which improves the efficiency of employee policy confirmation work and increases work productivity.

[0241] Example 2

[0242] 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."

[0243] Conventional information provision systems have had the problem that answers to user questions do not take emotions into consideration, resulting in low user satisfaction. Another problem is that it is difficult to accurately understand the content of the question entered by the user and provide a quick and appropriate answer. Furthermore, there is a need for systems that can provide more specific and useful information by generating answers that take past cases into account.

[0244] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for [analyzing the user's emotions using an emotion analysis engine], a means for [analyzing the question by natural language processing], and a means for [generating an answer based on the search results and formatting the answer by natural language generation while taking the emotions into consideration]. This makes it possible to provide answers that take the user's emotions into consideration and specific, useful information that takes past cases into consideration.

[0245] A "communication terminal" is a device used by a user to input a question and send it to a server, and includes a personal computer, smartphone, tablet, etc.

[0246] A "server" is an information processing device that includes a central device and software that receives questions from users, analyzes them, generates answers, and sends them to terminals.

[0247] An "emotion analysis engine" is a software module that analyzes a user's emotions from text entered by the user and identifies emotional states such as stress, anxiety, and joy.

[0248] "Natural language processing (NLP)" is a technology for analyzing questions entered by users and extracting keywords and the user's intent, and includes methods of text analysis and semantic understanding.

[0249] A "regulations database" is a database that stores employee regulations and company rules and guidelines.

[0250] A "case database" is a database that stores past cases and case studies.

[0251] "Natural language generation (NLG)" is a technology that generates sentences in a form that is easy for humans to understand based on search results and analytical information.

[0252] "Formatting with consideration for emotions" means creating a response format that is in line with the user's feelings by taking into account the user's emotional state and adding appropriate phrases to the response.

[0253] The present invention relates to an AI system that provides quick and appropriate answers to questions entered by a user through a terminal, and is characterized in that it generates answers by analyzing the user's emotions. The following describes in detail the embodiments of the present invention.

[0254] Users access the system using a communication device such as a PC or smartphone. A browser or dedicated application is installed on the communication device, and users can input and send questions. For example, suppose a user inputs a question such as, "Please tell me how to register attendance when visiting the hospital."

[0255] The device sends the questions entered by the user to the server using the HTTPS protocol, using SSL / TLS encryption technology to ensure a secure communication environment.

[0256] The server analyzes the questions received from the user. First, it uses a sentiment analysis engine to analyze the user's emotions. This sentiment analysis engine uses sentiment analysis APIs such as Azure Cognitive Services and IBM Watson. For example, if "anxiety" is detected, that information is used for subsequent processing.

[0257] The server then analyzes the question using a natural language processing (NLP) module, which utilizes technologies such as spaCy and TensorFlow. Specifically, it extracts keywords such as "hospital visit" and "attendance registration" from the question, as well as the user's intent.

[0258] The server then searches for relevant information from the policy and case databases based on the extracted keywords and intent. These databases are managed using database management systems such as MySQL and PostgreSQL. The search results include relevant information based on employee policy and past cases.

[0259] The server evaluates the search results and generates the most appropriate answer. This answer generation uses a natural language generation (NLG) module, such as OpenAI's GPT-3 or Google's BERT. At this stage, the results of the sentiment analysis engine are reflected, and the answer is formatted to take the user's emotions into consideration. For example, if the user is detected as "anxious," the answer may include phrases such as "Don't worry" or "Don't worry."

[0260] Finally, the server sends the generated answer to the device and displays it to the user. The user can check the specific answer on the screen of their communication device. For example, if the question is, "What is the procedure for taking paid leave?" and the emotion analysis engine detects "stress," the answer might include something like, "To take paid leave, please get your supervisor's approval and then apply through the attendance management system. It's okay, your health is important, so please get plenty of rest."

[0261] In this way, the system of the present invention can provide users with prompt and appropriate answers that take into consideration the user's feelings, and can provide users with accurate information based on employee regulations and past cases, which can be expected to deepen understanding of employee questions and improve work efficiency.

[0262] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0263] Step 1:

[0264] The user enters and sends a question from a communication device. The user accesses a specific web application or mobile application using a PC or smartphone and enters a question. For example, the user might enter, "Please tell me the procedure for taking paid leave." The entered question is sent to the server using the HTTPS protocol.

[0265] Input: User-entered question: "What is the procedure for taking paid leave?"

[0266] Output: The question data sent to the server

[0267] Step 2:

[0268] The device sends a query to the server. The device securely sends data to the server using SSL / TLS. The sent data is received by the server.

[0269] Input: Question data sent from the terminal

[0270] Output: The query data received by the server

[0271] Step 3:

[0272] The server analyzes the user's emotions using a sentiment analysis engine. The server inputs the received question data into the analysis engine and analyzes emotions (e.g., stress, anxiety, joy, etc.) from the question text. This analysis uses a sentiment analysis API (e.g., Azure Cognitive Services or IBM Watson).

[0273] Input: Received question data "Please tell me the procedure for taking paid leave."

[0274] Output: Emotion analysis results such as "stress"

[0275] Step 4:

[0276] The server analyzes the question using a natural language processing (NLP) module. The server passes the sentiment analysis results and question data to the NLP module, which extracts important keywords and user intent. NLP technologies (e.g., spaCy and TensorFlow) are used for this analysis.

[0277] Input: Question data "Please tell me the procedure for taking paid leave" and emotion analysis result "Stress"

[0278] Output: Keywords such as "paid leave," "acquisition," and "procedure" and user intent

[0279] Step 5:

[0280] The server searches for relevant information from the database. Based on the extracted keywords and intent, the server searches for information from the rule database and case database. At this stage, SQL queries are used to quickly obtain the required information.

[0281] Input: Keywords "paid leave", "acquisition", "procedure" and user intent

[0282] Output: Search results and past cases

[0283] Step 6:

[0284] The server generates answers and formats them taking emotions into consideration. The server inputs the search results into an NLG module, which generates answers in natural language based on the emotional information. For example, OpenAI GPT-3 and Google BERT are used in this process. Depending on the user's emotional state (e.g., stress), phrases such as "Don't worry" or "Don't worry" are added to the answer.

[0285] Input: Search results, past cases, and emotional information about "stress"

[0286] Output: Emotionally sensitive formatted answers

[0287] Step 7:

[0288] The server sends the generated response to the terminal. The server encodes the generated response in JSON format or similar and sends it to the terminal via HTTPS.

[0289] Input: Formatted answer

[0290] Output: Response data sent to the device

[0291] Step 8:

[0292] The terminal displays the answer to the user. The terminal decodes the received answer data into an appropriate format and displays it to the user. The user can check the specific answer on the terminal screen.

[0293] Input: Response data sent from the server

[0294] Output: The specific answer that is displayed to the user

[0295] The above is the specific processing flow of this system.

[0296] (Application example 2)

[0297] 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."

[0298] Conventional electronic payment systems have struggled to provide quick and appropriate answers to the various problems and questions users face. Furthermore, by returning mechanical answers without considering the user's emotional state, there is a risk of increasing the user's anxiety and stress. This can lead to a decline in user satisfaction and a deterioration in service quality. To solve this issue, a system is needed that provides answers that take the user's emotions into consideration and resolves problems quickly and accurately.

[0299] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for a user to input and send a question from a terminal, means for the server to receive the question and analyze the question using an emotion recognition engine and natural language processing, means for searching a database for rules and past cases based on the analysis results, means for generating an answer based on the search results and formatting it using natural language generation that reflects emotional information, and means for sending the generated answer to the terminal and displaying it to the user. This makes it possible to provide a quick and accurate answer while taking the user's emotional state into consideration, thereby improving user satisfaction and maintaining service quality.

[0300] The means by which "users can enter and submit questions from their devices" refers to an interface that allows users to access the system using devices such as smartphones or PCs, enter questions in text format, and submit them.

[0301] The means "the server receives the question and analyzes the question using an emotion recognition engine and natural language processing" refers to a process in which the server first analyzes the emotional state of the user's question received from the user using an emotion recognition engine, and then analyzes the content of the question using natural language processing.

[0302] The means for "searching for regulations and past cases from a database based on the analysis results" refers to a mechanism for searching for related regulation information and past cases from a database based on the analyzed question content and emotion information.

[0303] The method of "generating answers based on search results and formatting them using natural language generation that reflects emotional information" refers to the process of using the searched information to generate appropriate answers to the user's questions, and then formatting the answers in natural language using appropriate expressions, taking into account the user's emotional information.

[0304] The means for "sending the generated answer to the terminal and displaying it to the user" refers to a mechanism for sending the generated answer to the user's terminal and displaying it on the screen to provide it to the user.

[0305] The means of "selecting the most appropriate answer taking into consideration the user's emotional information" refers to the process of selecting and providing the answer from the search results that best corresponds to the user's emotional state.

[0306] The means of "using a regulations database and a database of past cases" refers to a mechanism for obtaining information from a regulations database in which organizational rules and guidelines are stored and a database in which similar past cases are stored.

[0307] The present invention relates to an AI system that recognizes a user's emotions and generates appropriate responses based on the emotions. Specific embodiments of the present invention will be described below.

[0308] First, a user accesses the system using a terminal (e.g., a smartphone or PC) and enters a question. For example, they might enter, "How do I change my credit card information?" This question is immediately sent from the terminal to the server.

[0309] The server analyzes the received question using an emotion recognition engine and a natural language processing (NLP) module. The emotion recognition engine analyzes emotions (e.g., anxiety, stress, joy, etc.) from the user's input text and obtains the results. For example, if the emotion engine detects "anxiety," that information is used in the subsequent analysis process.

[0310] The server then searches the rule database and case database based on the emotion information and the analyzed question content, thereby obtaining relevant information.

[0311] After retrieving the relevant information, the server generates an appropriate answer, taking into account the user's emotional state. A natural language generation (NLG) module assists this, formatting the answer in natural language to reflect the emotional information. For example, if the user is "feeling anxious" when asking a question, the server might generate an answer such as, "Don't worry, you can easily change your credit card information by following these steps."

[0312] Finally, the generated answer is sent to the terminal and displayed to the user, allowing the user to quickly obtain appropriate information for their question and receive an answer that takes their emotions into consideration.

[0313] Hardware and software examples

[0314] Hardware: Web server and user device (PC, smartphone)

[0315] software:

[0316] Emotion Recognition Engine: Hugging Face Transformers Library

[0317] Database: Regulation database and case database (e.g., PostgreSQL)

[0318] Natural Language Processing (NLP) module: uses OpenAI's GPT-3

[0319] Processing examples and prompts

[0320] Examples:

[0321] If a user asks "How do I change my credit card information?" and the emotion engine detects "anxiety," the server generates an answer that corresponds to the "anxious" state, such as "Don't worry, you can easily change your credit card information by following these steps."

[0322] Example prompt sentence:

[0323] Answer questions about electronic payment services when users feel uneasy. Question: How do I change my credit card information? Answer:

[0324] The present invention makes it possible to provide a prompt and appropriate response taking into account the user's emotional state, thereby improving user convenience and satisfaction.

[0325] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0326] Step 1:

[0327] The user inputs a question from the terminal and sends it.

[0328] The user enters a question using a smartphone or PC and presses the send button. For example, the user enters "How do I change my credit card information?" This input text is sent to the server.

[0329] Step 2:

[0330] The server receives the question and analyzes it using an emotion recognition engine and natural language processing.

[0331] The server processes the question received from the user. First, it uses an emotion recognition engine to analyze the sentiment of the text. In this case, "anxiety" is detected. Next, it uses a natural language processing module to analyze the keywords in the question and the user's intent. For example, the keywords "credit card information" and "change" are extracted.

[0332] Step 3:

[0333] Based on the analysis results, regulations and past cases are searched from the database.

[0334] The server then searches for relevant information from its database of company regulations and past case studies based on the analyzed keywords and intent. This operation retrieves regulations regarding changes to credit card information and similar past cases.

[0335] Step 4:

[0336] Answers are generated based on search results and formatted using natural language generation that reflects emotional information.

[0337] The server analyzes the retrieved information and generates an appropriate answer to the user's question. A natural language generation (NLG) module takes the user's sentiment into account to shape the answer. For example, the answer might be, "Don't worry, you can easily change your credit card information by following these steps."

[0338] Step 5:

[0339] The generated answer is sent to the terminal and displayed to the user.

[0340] The server generates a response and sends it to the user's device. The device receives the response and displays it to the user. The user can see the response on the device, such as "Don't worry, you can easily change your credit card information by following these steps."

[0341] 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.

[0342] 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.

[0343] 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.

[0344] [Second embodiment]

[0345] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0346] 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.

[0347] 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).

[0348] 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.

[0349] 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.

[0350] 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).

[0351] 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.

[0352] 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.

[0353] 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.

[0354] 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.

[0355] In the smart glasses 214, 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.

[0356] 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."

[0357] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases. The detailed implementation method for users to input questions and receive appropriate answers is described below.

[0358] First, the user accesses the system using a device such as a PC or smartphone and enters their question. For example, they might enter, "Please tell me how to register attendance when I go to the hospital." This question is immediately sent from the device to the server.

[0359] The server uses a natural language processing (NLP) module to analyze the questions received from the device. The server analyzes the questions and extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. This makes it clear which regulations should be referenced.

[0360] The server then searches the employee regulations database and past case database for related information based on the extracted keywords and intent, and retrieves the relevant employee regulations and similar past cases.

[0361] The server evaluates the search results and generates the most appropriate answer. The generated answer is formatted using a natural language generation (NLG) module to make it easy for the user to understand. For example, the server might generate an answer such as, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0362] Finally, the server sends the formatted answer to the terminal, and the terminal displays the generated answer to the user, allowing the user to quickly obtain appropriate information for the question, deepening their understanding of employee regulations and improving work efficiency.

[0363] As a concrete example, consider the case where a user asks, "What is the procedure for taking paid leave?" In this case, the server extracts keywords such as "paid leave," "taking," and "procedure," and searches for relevant regulations in the employee regulations database. It then takes into account similar past cases and generates an answer showing the appropriate procedure. For example, an answer including specific steps, such as "To take paid leave, please obtain approval from your supervisor and then apply through the attendance management system," is generated and presented to the user via their device.

[0364] As described above, the system of the present invention allows employees to get quick and accurate answers to their questions. This system significantly reduces the time and effort required to check employee regulations, thereby improving work efficiency.

[0365] The processing flow will be explained below.

[0366] Step 1:

[0367] User: Accesses the system and inputs a question into the interface on the terminal. For example, inputs "How do I register my attendance when I go to the hospital?"

[0368] Terminal: Receives user input and sends questions to the server.

[0369] Step 2:

[0370] Server: Receives questions sent as HTTP requests from the terminal.

[0371] Server: Passes the received question to a natural language processing (NLP) module to analyze the text.

[0372] Specific operation: Divide the question into tokens and extract keywords (e.g., "visiting the hospital" or "attendance registration") and intent.

[0373] Step 3:

[0374] Server: Based on the analysis results, a query is generated to search the employee regulations database and the past case database.

[0375] Specific operation: Generates SQL queries, etc., and retrieves information related to the question from the database.

[0376] Step 4:

[0377] Server: Aggregates information retrieved from the database and evaluates search results.

[0378] Specific operation: The system sorts the data and selects the most relevant information, such as "A doctor's note is required when visiting the hospital."

[0379] Step 5:

[0380] Server: Generates an answer to the user's question based on the evaluation results.

[0381] How it works: Responses are constructed based on employee policies and past experience, and formatted in a way that is easy for humans to understand. A natural language generation (NLG) module assists in this process.

[0382] Example: "When visiting a hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0383] Step 6:

[0384] Server: Sends the generated answer to the device.

[0385] Specific operation: The formatted response is returned to the terminal in JSON format or similar.

[0386] Step 7:

[0387] Terminal: Displays the answer received from the server on a user interface.

[0388] Specific behavior: Updates the UI to present the received information to the user in the appropriate format.

[0389] Step 8:

[0390] User: Check the answers displayed on the device and take action if necessary.

[0391] Example: Take specific action such as registering as "visiting the hospital" in the attendance management system.

[0392] Through these steps, users can get instant and accurate answers to their questions, streamlining the employee policy review process and improving work productivity.

[0393] Example 1

[0394] 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."

[0395] In modern companies, having employees be able to quickly and accurately refer to employee regulations and past cases directly leads to improved work efficiency. However, current systems require employees to manually search for regulations and cases, which is time-consuming and labor-intensive. To solve this problem, there is a need for a system that allows users to obtain appropriate answers simply by entering a question.

[0396] 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.

[0397] In this invention, the server includes a means for a user to input and send a query from a computer terminal, a means for the server to receive the query and analyze it using natural language processing, and a means for searching for regulations and past cases from a database based on the analysis results, thereby enabling a user to quickly and accurately refer to employee regulations and past cases.

[0398] A "user" is a person who operates a computer terminal to input a query and obtain a response from the system.

[0399] "Computer terminal" refers to a device that can connect to the Internet, such as a PC or smartphone.

[0400] A "query" is a question or command that a user enters into a system with the purpose of obtaining information.

[0401] A "server" is a computer system that receives inquiries, performs natural language processing and search processing, and generates and transmits answers.

[0402] "Natural language processing" is a technique used by computers to analyze human language and understand user queries.

[0403] "Analysis" is the process by which the server understands the content of the query and extracts relevant information.

[0404] A "database" is a collection of information in which information such as regulations and past cases is structured and stored.

[0405] "Policies" are documents or information that describe company rules and procedures that employees must follow.

[0406] "Past cases" are records of responses and actions taken in response to similar inquiries that have occurred in the past.

[0407] "Searching" is the process of finding specific information within a database.

[0408] A "response" is information or instructions that the server generates and sends in response to a user's inquiry.

[0409] "Generation" is the process by which the server creates a response for the user based on the analysis results.

[0410] "Natural language generation" is a technology that allows a computer to create text in a form that humans can understand.

[0411] "Transmission" means the act of electronically transferring a query or response from one party to another.

[0412] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases. Specific embodiments are described below.

[0413] Hardware and software used

[0414] This system mainly consists of the following hardware and software:

[0415] Hardware

[0416] 1. Computer terminal

[0417] A PC or smartphone for users to enter questions

[0418] 2. Server

[0419] A high-performance computer for analyzing questions, searching databases, and generating answers

[0420] software

[0421] 1. Natural Language Processing (NLP) Module

[0422] Software used to analyze user questions. Examples include NLP libraries such as spaCy and Transformers.

[0423] 2. Database Management System (DBMS)

[0424] It is used to execute SQL queries to search for employee policies and past cases. Examples include MySQL and PostgreSQL.

[0425] 3. Natural Language Generation (NLG) Module

[0426] It is used to generate answers based on search results and format them in a format that is easy for users to understand. Specific examples include OpenAI's GPT-3 and GPT-4.

[0427] Data processing and data calculation

[0428] 1. Question analysis

[0429] The question entered by the user on the computer terminal is sent to the server, which uses an NLP module to extract keywords such as "hospital visit" and "attendance registration" from the question and the user's intention.

[0430] 2. Database search

[0431] The server searches for relevant information from the employee regulations database and past case database based on the extracted keywords. This search uses SQL queries. For example, the SQL query for the keyword "hospital visit" is executed as follows: "SELECT FROM employee_regulations WHERE keyword='hospital visit'".

[0432] 3. Answer generation

[0433] The server evaluates the search results and generates the most appropriate answer using an NLG module, which is formatted in a way that is easy for the user to understand, such as "To take paid leave, please get your supervisor's approval and apply through the attendance management system."

[0434] Specific examples

[0435] A specific example will be given below in which a user asks on a computer terminal, "Please tell me how to register attendance when visiting a hospital."

[0436] 1. Enter your question

[0437] Users access the system using a PC or smartphone and enter their questions.

[0438] 2. Submit a question

[0439] The terminal immediately sends the query to the server.

[0440] 3. Question analysis

[0441] The server uses an NLP module to analyze the question and extract the keywords "hospital visit" and "attendance registration."

[0442] 4. Database Search

[0443] The server searches the employee regulations database and past case database for relevant information and acquires the necessary information.

[0444] 5. Answer generation

[0445] Based on the search results, the server uses the NLG module to generate a response such as, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0446] 6. Submit your response

[0447] The server sends the generated answer to the computer terminal, which displays it to the user.

[0448] Prompt Sentence Examples

[0449] "Please tell me how to register attendance when visiting the hospital."

[0450] What is the procedure for taking paid leave?

[0451] "I would like to know how to claim expenses when working from home."

[0452] By entering the above prompts, the user can quickly and accurately obtain information on employee regulations and past cases. The system of the present invention provides prompt and appropriate answers to the user's questions, significantly improving the work efficiency of employees.

[0453] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0454] Step 1:

[0455] A user accesses the system using a computer terminal and inputs a question. Specifically, the user opens a browser, accesses the system's web interface, and inputs a question into the input form. For example, the user might input, "Please tell me how to register attendance when visiting the hospital." This question becomes input data.

[0456] Step 2:

[0457] The terminal immediately sends the question entered by the user to the server. Specifically, when the user clicks the send button, the terminal sends the input data to the server as an HTTP POST request. This transmission becomes the input for processing.

[0458] Step 3:

[0459] The server analyzes the questions received from the device using a natural language processing (NLP) module. Specifically, the server tokenizes the questions received using an NLP library (e.g., spaCy or Transformers) and performs semantic analysis. This analysis extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. The analysis results are the output for data processing.

[0460] Step 4:

[0461] The server searches for relevant information from the employee regulations database and past case database based on the analysis results. Specifically, it generates an SQL query based on the extracted keywords and searches for data using a database management system such as MySQL or PostgreSQL. For example, an SQL query such as "SELECT FROM employee_regulations WHERE keyword='hospital visit'" is executed. The data retrieved by this search becomes the output of data calculations based on the input data.

[0462] Step 5:

[0463] The server evaluates the search results and generates the most appropriate answer. Specifically, it uses an NLG module (for example, OpenAI's GPT-3) to generate natural language based on the search results, and creates an answer in a format that is easy for the user to understand. For example, the answer could be something like, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system." This generated answer becomes the output data.

[0464] Step 6:

[0465] The server sends the generated answer to the terminal. Specifically, it organizes the generated answer in JSON format and sends it to the terminal as an HTTP response. This transmission becomes the output of the processing.

[0466] Step 7:

[0467] The device displays the answer received from the server to the user. Specifically, it updates the DOM of the web page and reflects the received text in the answer display area. As a result, the user can see the appropriate information for the question on the screen. This display is the final output.

[0468] (Application example 1)

[0469] 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."

[0470] In brick-and-mortar stores, when store clerks or managers have questions about store operations, it can be difficult to quickly and accurately obtain the appropriate information. For example, when they need to immediately check product return policies or complaint handling procedures, not being able to obtain the appropriate information in a timely manner can reduce work efficiency. Another issue is the lack of a way to easily check information on a device at hand while working on other tasks.

[0471] 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.

[0472] In this invention, the server includes means for a user to input and send a question from a terminal, means for the server to receive the question and analyze the question using natural language processing, means for searching a database for regulations and past cases based on the analysis results, means for generating an answer based on the search results and formatting it using natural language generation, and means for transmitting the generated answer to the terminal and displaying it to the user, thereby enabling users to quickly and accurately obtain appropriate information for questions about store operation in a physical store.

[0473] "Means for users to input and send questions from a terminal" refers to the process and interface by which a user inputs a specific question using a smart device and sends the question to a server over a network.

[0474] "Means for the server to receive a question and analyze the question using natural language processing" refers to a module or algorithm that allows the server to receive a question sent by a user, analyze the question using natural language processing technology, and understand its meaning and intent.

[0475] "Means for searching for regulations and past cases from a database based on the analysis results" refers to functions and modules that extract keywords from the results of the analyzed questions and use them to search a database in which regulations and past cases are stored.

[0476] "Means of generating answers based on search results and formatting them using natural language generation" refers to the function of using natural language generation technology to create answers in a format that is easy for users to understand based on information obtained from a database, and then formatting them.

[0477] "Means for transmitting the generated answer to the terminal and displaying it to the user" refers to the processes and techniques for transmitting the formatted answer back over the network to the originating smart device and visually displaying it to the user.

[0478] "A means of providing work support by displaying generated answers on a smart device" refers to a function that displays generated answers on smart devices such as smartphones, tablets, and smart glasses, and supports work related to store operations.

[0479] "Means of utilizing regulations related to store operations and databases of past cases" refers to the ability to utilize databases that record regulations related to store operations (such as refund policies and complaint handling procedures) and specific past cases.

[0480] This invention provides a question-answering system that uses AI technology to improve the efficiency of brick-and-mortar store operations. Specific embodiments of this system are described below.

[0481] First, a user (a store clerk or manager) uses a smart device (smartphone, tablet, smart glasses, etc.) to input and send a question. For example, they might input, "Please tell me the procedure for returning a product." This question is immediately sent to the server via the network.

[0482] The server uses a natural language processing (NLP) module such as Spacy to analyze the received question. Specifically, it extracts keywords such as "return" and "procedure" from the question and understands the user's intent.

[0483] The server then searches a database for relevant regulations and past cases based on the extracted keywords. The regulation database includes product return policies and store operation rules, while the past case database stores specific past cases. This allows for quick retrieval of the relevant information.

[0484] The server then generates an answer based on the information it has obtained, using natural language generation (NLG) modules like Hugging Face Transformers to format the answer in a way that is easy for the user to understand, such as specific instructions like "When returning an item, please bring the item with the receipt and fill out the return form."

[0485] The generated answer is sent to the original smart device via the network, where the user can visually check it and take any necessary action immediately, significantly improving the efficiency of store operations.

[0486] For example:

[0487] When a user asks, "What is the procedure for returning a product?", the server extracts keywords such as "return" and "procedure" and searches the database for relevant regulations and past cases. It then generates a response such as, "When returning a product, please bring the product along with the receipt and fill out the necessary information on the return form," which is displayed on the smart device. This process allows the user to take prompt and appropriate action.

[0488] Example prompt sentence:

[0489] Q: What is the procedure for returning a product?

[0490] A: When returning an item, please bring the item with you along with your receipt and fill out the return form.

[0491] This system makes it possible to quickly resolve questions about store operations and improve operational efficiency.

[0492] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0493] Step 1:

[0494] A user uses a smart device (smartphone, tablet, smart glasses, etc.) to input a question and press the "Send" button. For example, the user might input "Please tell me the procedure for returning a product." At this time, the input question is sent as text data from the device to the server.

[0495] Input: The question text entered by the user (e.g., "How do I return an item?")

[0496] Output: Submitted question text

[0497] Step 2:

[0498] The server receives the question sent from the device and analyzes it using a natural language processing (NLP) module that utilizes Spacy. Specifically, the question is tokenized and keywords such as "return" and "procedure" are extracted. The user's intent is understood based on the extracted keywords.

[0499] Input: Submitted question text

[0500] Output: Extracted keywords and user intent (e.g., "return," "procedure")

[0501] Step 3:

[0502] The server searches the regulatory database and past case database based on the extracted keywords, thereby obtaining relevant regulatory information on store operations and past case studies that have previously addressed similar questions. The search results are stored in list format.

[0503] Input: Extracted keywords and user intent

[0504] Output: Search result list (e.g. related regulatory information and past cases)

[0505] Step 4:

[0506] Based on the search results, the server uses a natural language generation (NLG) module that utilizes Hugging Face Transformers to generate answers in a format that is easy for the user to understand, specifically outputting answers in natural-sounding sentences while taking into account the context of the search results.

[0507] Input: Search result list

[0508] Output: The generated answer (e.g., "When returning an item, please bring the item with you along with your receipt and fill out the return form.")

[0509] Step 5:

[0510] The server sends the generated answer to the original smart device via the network, and the smart device displays the received answer on its interface so that the user can visually confirm it.

[0511] Input: Generated Answer

[0512] Output: Answer displayed on the user's terminal

[0513] Step 6:

[0514] The user checks the answers displayed on the smart device and takes specific action, such as actually returning the product based on the answers.

[0515] Input: Answer displayed on smart device

[0516] Output: User's actual action (e.g., returning a product)

[0517] In this way, by responding quickly and accurately to questions about running a physical store, the operational efficiency of the entire store can be improved.

[0518] 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.

[0519] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases, and by combining it with an emotion engine that recognizes the user's emotions, it is possible to obtain more appropriate answers. Specific embodiments of the present invention are described below.

[0520] First, the user accesses the system using a device such as a PC or smartphone and enters a question. For example, they might enter, "Please tell me how to register attendance when I go to the hospital." This question is immediately sent from the device to the server.

[0521] The server first uses an emotion engine to analyze the question received from the terminal. The emotion engine analyzes emotions (e.g., stress, anxiety, anger, joy, etc.) from the user's input text and obtains the results. For example, if the emotion engine detects "anxiety" from the user's question, that information is used in the subsequent analysis process.

[0522] The server then passes the acquired emotional information along with the question to a natural language processing (NLP) module. The server analyzes the question and extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. This makes it clear which regulations should be referenced.

[0523] The server searches the employee regulations database and past case database for related information based on the extracted keywords and intent, and obtains the relevant employee regulations and similar past cases.

[0524] The server evaluates the search results and generates the most appropriate answer. The generated answer is adjusted to reflect the user's emotional information. For example, if the user is feeling anxious when asking a question, the answer will include phrases such as "Don't worry" or "Don't worry," providing an answer that takes the user's emotions into consideration. This is supported by a natural language generation (NLG) module.

[0525] Finally, the server sends the formatted answer to the terminal, and the terminal displays the generated answer to the user. This allows the user to quickly obtain appropriate information for the question and receive an answer that takes into consideration their feelings, thereby deepening their understanding of employee regulations and improving work efficiency.

[0526] As a concrete example, consider the case where a user asks, "What is the procedure for taking paid leave?" and the emotion engine detects "stress." The server extracts keywords such as "paid leave," "taking," and "procedure," and searches for relevant regulations in the employee regulations database. It then generates an answer showing the appropriate procedure, taking into account similar past cases. For example, an answer that takes specific steps and emotions into consideration, such as "To take paid leave, please get your supervisor's approval and then apply through the attendance management system. It's okay, your health is important, so please take plenty of rest," is generated and presented to the user via their device.

[0527] As described above, by recognizing the user's emotions and adjusting responses accordingly, the system of the present invention not only allows employees to get quick and accurate answers to their questions, but also allows for more user-friendly responses by taking emotions into consideration, thereby streamlining the confirmation process for employee regulations and improving work productivity.

[0528] The processing flow will be explained below.

[0529] Step 1:

[0530] User: Accesses the system using a device such as a PC or smartphone and enters a question. For example, "Please tell me how to register attendance when going to the hospital."

[0531] Terminal: Receives user input and sends questions to the server.

[0532] Step 2:

[0533] Server: Receives questions sent as HTTP requests from the terminal.

[0534] Server: Passes the received question to the emotion engine and analyzes the user's emotion from the text.

[0535] Specific operation: Analyzes the question and applies an emotion model to extract emotions such as "stress," "anxiety," and "joy." For example, the emotion "anxiety" can be detected from the question "Please tell me how to register attendance when going to the hospital."

[0536] Step 3:

[0537] Server: Holds the extracted sentiment information and passes the question to a natural language processing (NLP) module to analyze the text.

[0538] Specific operation: The question is divided into tokens, and keywords (e.g., "visiting the hospital" or "attendance registration") and the user's intent are extracted.

[0539] Step 4:

[0540] Server: Based on the analysis results, a query is generated to search the employee regulations database and the past case database.

[0541] Specific operation: Generates SQL queries, etc., and retrieves information related to the question from the database.

[0542] Step 5:

[0543] Server: Aggregates information retrieved from the database and evaluates search results.

[0544] Specific behavior: Orders the retrieved data and identifies the most relevant information, such as "A doctor's note is required when visiting the hospital."

[0545] Step 6:

[0546] Server: Based on the evaluation results, it generates an answer to the user's question. It adjusts the answer taking into account emotional information.

[0547] How it works: The server constructs a response based on employee policy and past case data, and a natural language generation (NLG) module formats it. For example, if anxiety is detected, a phrase like "Don't worry" is added to the response.

[0548] Example: "When you visit the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system. Don't worry, everything will be fine."

[0549] Step 7:

[0550] Server: Sends the generated answer to the device.

[0551] Specific operation: The formatted response is returned to the terminal in JSON format or similar.

[0552] Step 8:

[0553] Terminal: Displays the answer received from the server on a user interface.

[0554] Specific behavior: Updates the UI to present the received information to the user in the appropriate format.

[0555] Step 9:

[0556] User: Check the answers displayed on the device and take action if necessary.

[0557] Example: Take specific action such as registering as "visiting the hospital" in the attendance management system.

[0558] This allows users to receive instant, accurate answers to their questions, and also allows them to receive answers that take into consideration their feelings, which improves the efficiency of employee policy confirmation work and increases work productivity.

[0559] Example 2

[0560] 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."

[0561] Conventional information provision systems have had the problem that answers to user questions do not take emotions into consideration, resulting in low user satisfaction. Another problem is that it is difficult to accurately understand the content of the question entered by the user and provide a quick and appropriate answer. Furthermore, there is a need for systems that can provide more specific and useful information by generating answers that take past cases into account.

[0562] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for [analyzing the user's emotions using an emotion analysis engine], a means for [analyzing the question by natural language processing], and a means for [generating an answer based on the search results and formatting the answer by natural language generation while taking the emotions into consideration]. This makes it possible to provide answers that take the user's emotions into consideration and specific, useful information that takes past cases into consideration.

[0563] A "communication terminal" is a device used by a user to input a question and send it to a server, and includes a personal computer, smartphone, tablet, etc.

[0564] A "server" is an information processing device that includes a central device and software that receives questions from users, analyzes them, generates answers, and sends them to terminals.

[0565] An "emotion analysis engine" is a software module that analyzes a user's emotions from text entered by the user and identifies emotional states such as stress, anxiety, and joy.

[0566] "Natural language processing (NLP)" is a technology for analyzing questions entered by users and extracting keywords and the user's intent, and includes methods of text analysis and semantic understanding.

[0567] A "regulations database" is a database that stores employee regulations and company rules and guidelines.

[0568] A "case database" is a database that stores past cases and case studies.

[0569] "Natural language generation (NLG)" is a technology that generates sentences in a form that is easy for humans to understand based on search results and analytical information.

[0570] "Formatting with consideration for emotions" means creating a response format that is in line with the user's feelings by taking into account the user's emotional state and adding appropriate phrases to the response.

[0571] The present invention relates to an AI system that provides quick and appropriate answers to questions entered by a user through a terminal, and is characterized in that it generates answers by analyzing the user's emotions. The following describes in detail the embodiments of the present invention.

[0572] Users access the system using a communication device such as a PC or smartphone. A browser or dedicated application is installed on the communication device, and users can input and send questions. For example, suppose a user inputs a question such as, "Please tell me how to register attendance when visiting the hospital."

[0573] The device sends the questions entered by the user to the server using the HTTPS protocol, using SSL / TLS encryption technology to ensure a secure communication environment.

[0574] The server analyzes the questions received from the user. First, it uses a sentiment analysis engine to analyze the user's emotions. This sentiment analysis engine uses sentiment analysis APIs such as Azure Cognitive Services and IBM Watson. For example, if "anxiety" is detected, that information is used for subsequent processing.

[0575] The server then analyzes the question using a natural language processing (NLP) module, which utilizes technologies such as spaCy and TensorFlow. Specifically, it extracts keywords such as "hospital visit" and "attendance registration" from the question, as well as the user's intent.

[0576] The server then searches for relevant information from the policy and case databases based on the extracted keywords and intent. These databases are managed using database management systems such as MySQL and PostgreSQL. The search results include relevant information based on employee policy and past cases.

[0577] The server evaluates the search results and generates the most appropriate answer. This answer generation uses a natural language generation (NLG) module, such as OpenAI's GPT-3 or Google's BERT. At this stage, the results of the sentiment analysis engine are reflected, and the answer is formatted to take the user's emotions into consideration. For example, if the user is detected as "anxious," the answer may include phrases such as "Don't worry" or "Don't worry."

[0578] Finally, the server sends the generated answer to the device and displays it to the user. The user can check the specific answer on the screen of their communication device. For example, if the question is, "What is the procedure for taking paid leave?" and the emotion analysis engine detects "stress," the answer might include something like, "To take paid leave, please get your supervisor's approval and then apply through the attendance management system. It's okay, your health is important, so please get plenty of rest."

[0579] In this way, the system of the present invention can provide users with prompt and appropriate answers that take into consideration the user's feelings, and can provide users with accurate information based on employee regulations and past cases, which can be expected to deepen understanding of employee questions and improve work efficiency.

[0580] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0581] Step 1:

[0582] The user enters and sends a question from a communication device. The user accesses a specific web application or mobile application using a PC or smartphone and enters a question. For example, the user might enter, "Please tell me the procedure for taking paid leave." The entered question is sent to the server using the HTTPS protocol.

[0583] Input: User-entered question: "What is the procedure for taking paid leave?"

[0584] Output: The question data sent to the server

[0585] Step 2:

[0586] The device sends a query to the server. The device securely sends data to the server using SSL / TLS. The sent data is received by the server.

[0587] Input: Question data sent from the terminal

[0588] Output: The query data received by the server

[0589] Step 3:

[0590] The server analyzes the user's emotions using a sentiment analysis engine. The server inputs the received question data into the analysis engine and analyzes emotions (e.g., stress, anxiety, joy, etc.) from the question text. This analysis uses a sentiment analysis API (e.g., Azure Cognitive Services or IBM Watson).

[0591] Input: Received question data "Please tell me the procedure for taking paid leave."

[0592] Output: Emotion analysis results such as "stress"

[0593] Step 4:

[0594] The server analyzes the question using a natural language processing (NLP) module. The server passes the sentiment analysis results and question data to the NLP module, which extracts important keywords and user intent. NLP technologies (e.g., spaCy and TensorFlow) are used for this analysis.

[0595] Input: Question data "Please tell me the procedure for taking paid leave" and emotion analysis result "Stress"

[0596] Output: Keywords such as "paid leave," "acquisition," and "procedure" and user intent

[0597] Step 5:

[0598] The server searches for relevant information from the database. Based on the extracted keywords and intent, the server searches for information from the rule database and case database. At this stage, SQL queries are used to quickly obtain the required information.

[0599] Input: Keywords "paid leave", "acquisition", "procedure" and user intent

[0600] Output: Search results and past cases

[0601] Step 6:

[0602] The server generates answers and formats them taking emotions into consideration. The server inputs the search results into an NLG module, which generates answers in natural language based on the emotional information. For example, OpenAI GPT-3 and Google BERT are used in this process. Depending on the user's emotional state (e.g., stress), phrases such as "Don't worry" or "Don't worry" are added to the answer.

[0603] Input: Search results, past cases, and emotional information about "stress"

[0604] Output: Emotionally sensitive formatted answers

[0605] Step 7:

[0606] The server sends the generated response to the terminal. The server encodes the generated response in JSON format or similar and sends it to the terminal via HTTPS.

[0607] Input: Formatted answer

[0608] Output: Response data sent to the device

[0609] Step 8:

[0610] The terminal displays the answer to the user. The terminal decodes the received answer data into an appropriate format and displays it to the user. The user can check the specific answer on the terminal screen.

[0611] Input: Response data sent from the server

[0612] Output: The specific answer that is displayed to the user

[0613] The above is the specific processing flow of this system.

[0614] (Application example 2)

[0615] 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."

[0616] Conventional electronic payment systems have struggled to provide quick and appropriate answers to the various problems and questions users face. Furthermore, by returning mechanical answers without considering the user's emotional state, there is a risk of increasing the user's anxiety and stress. This can lead to a decline in user satisfaction and a deterioration in service quality. To solve this issue, a system is needed that provides answers that take the user's emotions into consideration and resolves problems quickly and accurately.

[0617] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for a user to input and send a question from a terminal, means for the server to receive the question and analyze the question using an emotion recognition engine and natural language processing, means for searching a database for rules and past cases based on the analysis results, means for generating an answer based on the search results and formatting it using natural language generation that reflects emotional information, and means for sending the generated answer to the terminal and displaying it to the user. This makes it possible to provide a quick and accurate answer while taking the user's emotional state into consideration, thereby improving user satisfaction and maintaining service quality.

[0618] The means by which "users can enter and submit questions from their devices" refers to an interface that allows users to access the system using devices such as smartphones or PCs, enter questions in text format, and submit them.

[0619] The means "the server receives the question and analyzes the question using an emotion recognition engine and natural language processing" refers to a process in which the server first analyzes the emotional state of the user's question received from the user using an emotion recognition engine, and then analyzes the content of the question using natural language processing.

[0620] The means for "searching for regulations and past cases from a database based on the analysis results" refers to a mechanism for searching for related regulation information and past cases from a database based on the analyzed question content and emotion information.

[0621] The method of "generating answers based on search results and formatting them using natural language generation that reflects emotional information" refers to the process of using the searched information to generate appropriate answers to the user's questions, and then formatting the answers in natural language using appropriate expressions, taking into account the user's emotional information.

[0622] The means for "sending the generated answer to the terminal and displaying it to the user" refers to a mechanism for sending the generated answer to the user's terminal and displaying it on the screen to provide it to the user.

[0623] The means of "selecting the most appropriate answer taking into consideration the user's emotional information" refers to the process of selecting and providing the answer from the search results that best corresponds to the user's emotional state.

[0624] The means of "using a regulations database and a database of past cases" refers to a mechanism for obtaining information from a regulations database in which organizational rules and guidelines are stored and a database in which similar past cases are stored.

[0625] The present invention relates to an AI system that recognizes a user's emotions and generates appropriate responses based on the emotions. Specific embodiments of the present invention will be described below.

[0626] First, a user accesses the system using a terminal (e.g., a smartphone or PC) and enters a question. For example, they might enter, "How do I change my credit card information?" This question is immediately sent from the terminal to the server.

[0627] The server analyzes the received question using an emotion recognition engine and a natural language processing (NLP) module. The emotion recognition engine analyzes emotions (e.g., anxiety, stress, joy, etc.) from the user's input text and obtains the results. For example, if the emotion engine detects "anxiety," that information is used in the subsequent analysis process.

[0628] The server then searches the rule database and case database based on the emotion information and the analyzed question content, thereby obtaining relevant information.

[0629] After retrieving the relevant information, the server generates an appropriate answer, taking into account the user's emotional state. A natural language generation (NLG) module assists this, formatting the answer in natural language to reflect the emotional information. For example, if the user is "feeling anxious" when asking a question, the server might generate an answer such as, "Don't worry, you can easily change your credit card information by following these steps."

[0630] Finally, the generated answer is sent to the terminal and displayed to the user, allowing the user to quickly obtain appropriate information for their question and receive an answer that takes their emotions into consideration.

[0631] Hardware and software examples

[0632] Hardware: Web server and user device (PC, smartphone)

[0633] software:

[0634] Emotion Recognition Engine: Hugging Face Transformers Library

[0635] Database: Regulation database and case database (e.g., PostgreSQL)

[0636] Natural Language Processing (NLP) module: uses OpenAI's GPT-3

[0637] Processing examples and prompts

[0638] Examples:

[0639] If a user asks "How do I change my credit card information?" and the emotion engine detects "anxiety," the server generates an answer that corresponds to the "anxious" state, such as "Don't worry, you can easily change your credit card information by following these steps."

[0640] Example prompt sentence:

[0641] Answer questions about electronic payment services when users feel uneasy. Question: How do I change my credit card information? Answer:

[0642] The present invention makes it possible to provide a prompt and appropriate response taking into account the user's emotional state, thereby improving user convenience and satisfaction.

[0643] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0644] Step 1:

[0645] The user inputs a question from the terminal and sends it.

[0646] The user enters a question using a smartphone or PC and presses the send button. For example, the user enters "How do I change my credit card information?" This input text is sent to the server.

[0647] Step 2:

[0648] The server receives the question and analyzes it using an emotion recognition engine and natural language processing.

[0649] The server processes the question received from the user. First, it uses an emotion recognition engine to analyze the sentiment of the text. In this case, "anxiety" is detected. Next, it uses a natural language processing module to analyze the keywords in the question and the user's intent. For example, the keywords "credit card information" and "change" are extracted.

[0650] Step 3:

[0651] Based on the analysis results, regulations and past cases are searched from the database.

[0652] The server then searches for relevant information from its database of company regulations and past case studies based on the analyzed keywords and intent. This operation retrieves regulations regarding changes to credit card information and similar past cases.

[0653] Step 4:

[0654] Answers are generated based on search results and formatted using natural language generation that reflects emotional information.

[0655] The server analyzes the retrieved information and generates an appropriate answer to the user's question. A natural language generation (NLG) module takes the user's sentiment into account to shape the answer. For example, the answer might be, "Don't worry, you can easily change your credit card information by following these steps."

[0656] Step 5:

[0657] The generated answer is sent to the terminal and displayed to the user.

[0658] The server generates a response and sends it to the user's device. The device receives the response and displays it to the user. The user can see the response on the device, such as "Don't worry, you can easily change your credit card information by following these steps."

[0659] 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.

[0660] 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.

[0661] 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.

[0662] [Third embodiment]

[0663] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0664] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0665] 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).

[0666] 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.

[0667] 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.

[0668] 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).

[0669] 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.

[0670] 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.

[0671] 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.

[0672] 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.

[0673] 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.

[0674] 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."

[0675] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases. The detailed implementation method for users to input questions and receive appropriate answers is described below.

[0676] First, the user accesses the system using a device such as a PC or smartphone and enters their question. For example, they might enter, "Please tell me how to register attendance when I go to the hospital." This question is immediately sent from the device to the server.

[0677] The server uses a natural language processing (NLP) module to analyze the questions received from the device. The server analyzes the questions and extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. This makes it clear which regulations should be referenced.

[0678] The server then searches the employee regulations database and past case database for related information based on the extracted keywords and intent, and retrieves the relevant employee regulations and similar past cases.

[0679] The server evaluates the search results and generates the most appropriate answer. The generated answer is formatted using a natural language generation (NLG) module to make it easy for the user to understand. For example, the server might generate an answer such as, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0680] Finally, the server sends the formatted answer to the terminal, and the terminal displays the generated answer to the user, allowing the user to quickly obtain appropriate information for the question, deepening their understanding of employee regulations and improving work efficiency.

[0681] As a concrete example, consider the case where a user asks, "What is the procedure for taking paid leave?" In this case, the server extracts keywords such as "paid leave," "taking," and "procedure," and searches for relevant regulations in the employee regulations database. It then takes into account similar past cases and generates an answer showing the appropriate procedure. For example, an answer including specific steps, such as "To take paid leave, please obtain approval from your supervisor and then apply through the attendance management system," is generated and presented to the user via their device.

[0682] As described above, the system of the present invention allows employees to get quick and accurate answers to their questions. This system significantly reduces the time and effort required to check employee regulations, thereby improving work efficiency.

[0683] The processing flow will be explained below.

[0684] Step 1:

[0685] User: Accesses the system and inputs a question into the interface on the terminal. For example, inputs "How do I register my attendance when I go to the hospital?"

[0686] Terminal: Receives user input and sends questions to the server.

[0687] Step 2:

[0688] Server: Receives questions sent as HTTP requests from the terminal.

[0689] Server: Passes the received question to a natural language processing (NLP) module to analyze the text.

[0690] Specific operation: Divide the question into tokens and extract keywords (e.g., "visiting the hospital" or "attendance registration") and intent.

[0691] Step 3:

[0692] Server: Based on the analysis results, a query is generated to search the employee regulations database and the past case database.

[0693] Specific operation: Generates SQL queries, etc., and retrieves information related to the question from the database.

[0694] Step 4:

[0695] Server: Aggregates information retrieved from the database and evaluates search results.

[0696] Specific operation: The system sorts the data and selects the most relevant information, such as "A doctor's note is required when visiting the hospital."

[0697] Step 5:

[0698] Server: Generates an answer to the user's question based on the evaluation results.

[0699] How it works: Responses are constructed based on employee policies and past experience, and formatted in a way that is easy for humans to understand. A natural language generation (NLG) module assists in this process.

[0700] Example: "When visiting a hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0701] Step 6:

[0702] Server: Sends the generated answer to the device.

[0703] Specific operation: The formatted response is returned to the terminal in JSON format or similar.

[0704] Step 7:

[0705] Terminal: Displays the answer received from the server on a user interface.

[0706] Specific behavior: Updates the UI to present the received information to the user in the appropriate format.

[0707] Step 8:

[0708] User: Check the answers displayed on the device and take action if necessary.

[0709] Example: Take specific action such as registering as "visiting the hospital" in the attendance management system.

[0710] Through these steps, users can get instant and accurate answers to their questions, streamlining the employee policy review process and improving work productivity.

[0711] Example 1

[0712] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0713] In modern companies, having employees be able to quickly and accurately refer to employee regulations and past cases directly leads to improved work efficiency. However, current systems require employees to manually search for regulations and cases, which is time-consuming and labor-intensive. To solve this problem, there is a need for a system that allows users to obtain appropriate answers simply by entering a question.

[0714] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0715] In this invention, the server includes a means for a user to input and send a query from a computer terminal, a means for the server to receive the query and analyze it using natural language processing, and a means for searching for regulations and past cases from a database based on the analysis results, thereby enabling a user to quickly and accurately refer to employee regulations and past cases.

[0716] A "user" is a person who operates a computer terminal to input a query and obtain a response from the system.

[0717] "Computer terminal" refers to a device that can connect to the Internet, such as a PC or smartphone.

[0718] A "query" is a question or command that a user enters into a system with the purpose of obtaining information.

[0719] A "server" is a computer system that receives inquiries, performs natural language processing and search processing, and generates and transmits answers.

[0720] "Natural language processing" is a technique used by computers to analyze human language and understand user queries.

[0721] "Analysis" is the process by which the server understands the content of the query and extracts relevant information.

[0722] A "database" is a collection of information in which information such as regulations and past cases is structured and stored.

[0723] "Policies" are documents or information that describe company rules and procedures that employees must follow.

[0724] "Past cases" are records of responses and actions taken in response to similar inquiries that have occurred in the past.

[0725] "Searching" is the process of finding specific information within a database.

[0726] A "response" is information or instructions that the server generates and sends in response to a user's inquiry.

[0727] "Generation" is the process by which the server creates a response for the user based on the analysis results.

[0728] "Natural language generation" is a technology that allows a computer to create text in a form that humans can understand.

[0729] "Transmission" means the act of electronically transferring a query or response from one party to another.

[0730] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases. Specific embodiments are described below.

[0731] Hardware and software used

[0732] This system mainly consists of the following hardware and software:

[0733] Hardware

[0734] 1. Computer terminal

[0735] A PC or smartphone for users to enter questions

[0736] 2. Server

[0737] A high-performance computer for analyzing questions, searching databases, and generating answers

[0738] software

[0739] 1. Natural Language Processing (NLP) Module

[0740] Software used to analyze user questions. Examples include NLP libraries such as spaCy and Transformers.

[0741] 2. Database Management System (DBMS)

[0742] It is used to execute SQL queries to search for employee policies and past cases. Examples include MySQL and PostgreSQL.

[0743] 3. Natural Language Generation (NLG) Module

[0744] It is used to generate answers based on search results and format them in a format that is easy for users to understand. Specific examples include OpenAI's GPT-3 and GPT-4.

[0745] Data processing and data calculation

[0746] 1. Question analysis

[0747] The question entered by the user on the computer terminal is sent to the server, which uses an NLP module to extract keywords such as "hospital visit" and "attendance registration" from the question and the user's intention.

[0748] 2. Database search

[0749] The server searches for relevant information from the employee regulations database and past case database based on the extracted keywords. This search uses SQL queries. For example, the SQL query for the keyword "hospital visit" is executed as follows: "SELECT FROM employee_regulations WHERE keyword='hospital visit'".

[0750] 3. Answer generation

[0751] The server evaluates the search results and generates the most appropriate answer using an NLG module, which is formatted in a way that is easy for the user to understand, such as "To take paid leave, please get your supervisor's approval and apply through the attendance management system."

[0752] Specific examples

[0753] A specific example will be given below in which a user asks on a computer terminal, "Please tell me how to register attendance when visiting a hospital."

[0754] 1. Enter your question

[0755] Users access the system using a PC or smartphone and enter their questions.

[0756] 2. Submit a question

[0757] The terminal immediately sends the query to the server.

[0758] 3. Question analysis

[0759] The server uses an NLP module to analyze the question and extract the keywords "hospital visit" and "attendance registration."

[0760] 4. Database Search

[0761] The server searches the employee regulations database and past case database for relevant information and acquires the necessary information.

[0762] 5. Answer generation

[0763] Based on the search results, the server uses the NLG module to generate a response such as, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0764] 6. Submit your response

[0765] The server sends the generated answer to the computer terminal, which displays it to the user.

[0766] Prompt Sentence Examples

[0767] "Please tell me how to register attendance when visiting the hospital."

[0768] What is the procedure for taking paid leave?

[0769] "I would like to know how to claim expenses when working from home."

[0770] By entering the above prompts, the user can quickly and accurately obtain information on employee regulations and past cases. The system of the present invention provides prompt and appropriate answers to the user's questions, significantly improving the work efficiency of employees.

[0771] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0772] Step 1:

[0773] A user accesses the system using a computer terminal and inputs a question. Specifically, the user opens a browser, accesses the system's web interface, and inputs a question into the input form. For example, the user might input, "Please tell me how to register attendance when visiting the hospital." This question becomes input data.

[0774] Step 2:

[0775] The terminal immediately sends the question entered by the user to the server. Specifically, when the user clicks the send button, the terminal sends the input data to the server as an HTTP POST request. This transmission becomes the input for processing.

[0776] Step 3:

[0777] The server analyzes the questions received from the device using a natural language processing (NLP) module. Specifically, the server tokenizes the questions received using an NLP library (e.g., spaCy or Transformers) and performs semantic analysis. This analysis extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. The analysis results are the output for data processing.

[0778] Step 4:

[0779] The server searches for relevant information from the employee regulations database and past case database based on the analysis results. Specifically, it generates an SQL query based on the extracted keywords and searches for data using a database management system such as MySQL or PostgreSQL. For example, an SQL query such as "SELECT FROM employee_regulations WHERE keyword='hospital visit'" is executed. The data retrieved by this search becomes the output of data calculations based on the input data.

[0780] Step 5:

[0781] The server evaluates the search results and generates the most appropriate answer. Specifically, it uses an NLG module (for example, OpenAI's GPT-3) to generate natural language based on the search results, and creates an answer in a format that is easy for the user to understand. For example, the answer could be something like, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system." This generated answer becomes the output data.

[0782] Step 6:

[0783] The server sends the generated answer to the terminal. Specifically, it organizes the generated answer in JSON format and sends it to the terminal as an HTTP response. This transmission becomes the output of the processing.

[0784] Step 7:

[0785] The device displays the answer received from the server to the user. Specifically, it updates the DOM of the web page and reflects the received text in the answer display area. As a result, the user can see the appropriate information for the question on the screen. This display is the final output.

[0786] (Application example 1)

[0787] 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."

[0788] In brick-and-mortar stores, when store clerks or managers have questions about store operations, it can be difficult to quickly and accurately obtain the appropriate information. For example, when they need to immediately check product return policies or complaint handling procedures, not being able to obtain the appropriate information in a timely manner can reduce work efficiency. Another issue is the lack of a way to easily check information on a device at hand while working on other tasks.

[0789] 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.

[0790] In this invention, the server includes means for a user to input and send a question from a terminal, means for the server to receive the question and analyze the question using natural language processing, means for searching a database for regulations and past cases based on the analysis results, means for generating an answer based on the search results and formatting it using natural language generation, and means for transmitting the generated answer to the terminal and displaying it to the user, thereby enabling users to quickly and accurately obtain appropriate information for questions about store operation in a physical store.

[0791] "Means for users to input and send questions from a terminal" refers to the process and interface by which a user inputs a specific question using a smart device and sends the question to a server over a network.

[0792] "Means for the server to receive a question and analyze the question using natural language processing" refers to a module or algorithm that allows the server to receive a question sent by a user, analyze the question using natural language processing technology, and understand its meaning and intent.

[0793] "Means for searching for regulations and past cases from a database based on the analysis results" refers to functions and modules that extract keywords from the results of the analyzed questions and use them to search a database in which regulations and past cases are stored.

[0794] "Means of generating answers based on search results and formatting them using natural language generation" refers to the function of using natural language generation technology to create answers in a format that is easy for users to understand based on information obtained from a database, and then formatting them.

[0795] "Means for transmitting the generated answer to the terminal and displaying it to the user" refers to the processes and techniques for transmitting the formatted answer back over the network to the originating smart device and visually displaying it to the user.

[0796] "A means of providing work support by displaying generated answers on a smart device" refers to a function that displays generated answers on smart devices such as smartphones, tablets, and smart glasses, and supports work related to store operations.

[0797] "Means of utilizing regulations related to store operations and databases of past cases" refers to the ability to utilize databases that record regulations related to store operations (such as refund policies and complaint handling procedures) and specific past cases.

[0798] This invention provides a question-answering system that uses AI technology to improve the efficiency of brick-and-mortar store operations. Specific embodiments of this system are described below.

[0799] First, a user (a store clerk or manager) uses a smart device (smartphone, tablet, smart glasses, etc.) to input and send a question. For example, they might input, "Please tell me the procedure for returning a product." This question is immediately sent to the server via the network.

[0800] The server uses a natural language processing (NLP) module such as Spacy to analyze the received question. Specifically, it extracts keywords such as "return" and "procedure" from the question and understands the user's intent.

[0801] The server then searches a database for relevant regulations and past cases based on the extracted keywords. The regulation database includes product return policies and store operation rules, while the past case database stores specific past cases. This allows for quick retrieval of the relevant information.

[0802] The server then generates an answer based on the information it has obtained, using natural language generation (NLG) modules like Hugging Face Transformers to format the answer in a way that is easy for the user to understand, such as specific instructions like "When returning an item, please bring the item with the receipt and fill out the return form."

[0803] The generated answer is sent to the original smart device via the network, where the user can visually check it and take any necessary action immediately, significantly improving the efficiency of store operations.

[0804] For example:

[0805] When a user asks, "What is the procedure for returning a product?", the server extracts keywords such as "return" and "procedure" and searches the database for relevant regulations and past cases. It then generates a response such as, "When returning a product, please bring the product along with the receipt and fill out the necessary information on the return form," which is displayed on the smart device. This process allows the user to take prompt and appropriate action.

[0806] Example prompt sentence:

[0807] Q: What is the procedure for returning a product?

[0808] A: When returning an item, please bring the item with you along with your receipt and fill out the return form.

[0809] This system makes it possible to quickly resolve questions about store operations and improve operational efficiency.

[0810] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0811] Step 1:

[0812] A user uses a smart device (smartphone, tablet, smart glasses, etc.) to input a question and press the "Send" button. For example, the user might input "Please tell me the procedure for returning a product." At this time, the input question is sent as text data from the device to the server.

[0813] Input: The question text entered by the user (e.g., "How do I return an item?")

[0814] Output: Submitted question text

[0815] Step 2:

[0816] The server receives the question sent from the device and analyzes it using a natural language processing (NLP) module that utilizes Spacy. Specifically, the question is tokenized and keywords such as "return" and "procedure" are extracted. The user's intent is understood based on the extracted keywords.

[0817] Input: Submitted question text

[0818] Output: Extracted keywords and user intent (e.g., "return," "procedure")

[0819] Step 3:

[0820] The server searches the regulatory database and past case database based on the extracted keywords, thereby obtaining relevant regulatory information on store operations and past case studies that have previously addressed similar questions. The search results are stored in list format.

[0821] Input: Extracted keywords and user intent

[0822] Output: Search result list (e.g. related regulatory information and past cases)

[0823] Step 4:

[0824] Based on the search results, the server uses a natural language generation (NLG) module that utilizes Hugging Face Transformers to generate answers in a format that is easy for the user to understand, specifically outputting answers in natural-sounding sentences while taking into account the context of the search results.

[0825] Input: Search result list

[0826] Output: The generated answer (e.g., "When returning an item, please bring the item with you along with your receipt and fill out the return form.")

[0827] Step 5:

[0828] The server sends the generated answer to the original smart device via the network, and the smart device displays the received answer on its interface so that the user can visually confirm it.

[0829] Input: Generated Answer

[0830] Output: Answer displayed on the user's terminal

[0831] Step 6:

[0832] The user checks the answers displayed on the smart device and takes specific action, such as actually returning the product based on the answers.

[0833] Input: Answer displayed on smart device

[0834] Output: User's actual action (e.g., returning a product)

[0835] In this way, by responding quickly and accurately to questions about running a physical store, the operational efficiency of the entire store can be improved.

[0836] 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.

[0837] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases, and by combining it with an emotion engine that recognizes the user's emotions, it is possible to obtain more appropriate answers. Specific embodiments of the present invention are described below.

[0838] First, the user accesses the system using a device such as a PC or smartphone and enters a question. For example, they might enter, "Please tell me how to register attendance when I go to the hospital." This question is immediately sent from the device to the server.

[0839] The server first uses an emotion engine to analyze the question received from the terminal. The emotion engine analyzes emotions (e.g., stress, anxiety, anger, joy, etc.) from the user's input text and obtains the results. For example, if the emotion engine detects "anxiety" from the user's question, that information is used in the subsequent analysis process.

[0840] The server then passes the acquired emotional information along with the question to a natural language processing (NLP) module. The server analyzes the question and extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. This makes it clear which regulations should be referenced.

[0841] The server searches the employee regulations database and past case database for related information based on the extracted keywords and intent, and obtains the relevant employee regulations and similar past cases.

[0842] The server evaluates the search results and generates the most appropriate answer. The generated answer is adjusted to reflect the user's emotional information. For example, if the user is feeling anxious when asking a question, the answer will include phrases such as "Don't worry" or "Don't worry," providing an answer that takes the user's emotions into consideration. This is supported by a natural language generation (NLG) module.

[0843] Finally, the server sends the formatted answer to the terminal, and the terminal displays the generated answer to the user. This allows the user to quickly obtain appropriate information for the question and receive an answer that takes into consideration their feelings, thereby deepening their understanding of employee regulations and improving work efficiency.

[0844] As a concrete example, consider the case where a user asks, "What is the procedure for taking paid leave?" and the emotion engine detects "stress." The server extracts keywords such as "paid leave," "taking," and "procedure," and searches for relevant regulations in the employee regulations database. It then generates an answer showing the appropriate procedure, taking into account similar past cases. For example, an answer that takes specific steps and emotions into consideration, such as "To take paid leave, please get your supervisor's approval and then apply through the attendance management system. It's okay, your health is important, so please take plenty of rest," is generated and presented to the user via their device.

[0845] As described above, by recognizing the user's emotions and adjusting responses accordingly, the system of the present invention not only allows employees to get quick and accurate answers to their questions, but also allows for more user-friendly responses by taking emotions into consideration, thereby streamlining the confirmation process for employee regulations and improving work productivity.

[0846] The processing flow will be explained below.

[0847] Step 1:

[0848] User: Accesses the system using a device such as a PC or smartphone and enters a question. For example, "Please tell me how to register attendance when going to the hospital."

[0849] Terminal: Receives user input and sends questions to the server.

[0850] Step 2:

[0851] Server: Receives questions sent as HTTP requests from the terminal.

[0852] Server: Passes the received question to the emotion engine and analyzes the user's emotion from the text.

[0853] Specific operation: Analyzes the question and applies an emotion model to extract emotions such as "stress," "anxiety," and "joy." For example, the emotion "anxiety" can be detected from the question "Please tell me how to register attendance when going to the hospital."

[0854] Step 3:

[0855] Server: Holds the extracted sentiment information and passes the question to a natural language processing (NLP) module to analyze the text.

[0856] Specific operation: The question is divided into tokens, and keywords (e.g., "visiting the hospital" or "attendance registration") and the user's intent are extracted.

[0857] Step 4:

[0858] Server: Based on the analysis results, a query is generated to search the employee regulations database and the past case database.

[0859] Specific operation: Generates SQL queries, etc., and retrieves information related to the question from the database.

[0860] Step 5:

[0861] Server: Aggregates information retrieved from the database and evaluates search results.

[0862] Specific behavior: Orders the retrieved data and identifies the most relevant information, such as "A doctor's note is required when visiting the hospital."

[0863] Step 6:

[0864] Server: Based on the evaluation results, it generates an answer to the user's question. It adjusts the answer taking into account emotional information.

[0865] How it works: The server constructs a response based on employee policy and past case data, and a natural language generation (NLG) module formats it. For example, if anxiety is detected, a phrase like "Don't worry" is added to the response.

[0866] Example: "When you visit the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system. Don't worry, everything will be fine."

[0867] Step 7:

[0868] Server: Sends the generated answer to the device.

[0869] Specific operation: The formatted response is returned to the terminal in JSON format or similar.

[0870] Step 8:

[0871] Terminal: Displays the answer received from the server on a user interface.

[0872] Specific behavior: Updates the UI to present the received information to the user in the appropriate format.

[0873] Step 9:

[0874] User: Check the answers displayed on the device and take action if necessary.

[0875] Example: Take specific action such as registering as "visiting the hospital" in the attendance management system.

[0876] This allows users to receive instant, accurate answers to their questions, and also allows them to receive answers that take into consideration their feelings, which improves the efficiency of employee policy confirmation work and increases work productivity.

[0877] Example 2

[0878] 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."

[0879] Conventional information provision systems have had the problem that answers to user questions do not take emotions into consideration, resulting in low user satisfaction. Another problem is that it is difficult to accurately understand the content of the question entered by the user and provide a quick and appropriate answer. Furthermore, there is a need for systems that can provide more specific and useful information by generating answers that take past cases into account.

[0880] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for [analyzing the user's emotions using an emotion analysis engine], a means for [analyzing the question by natural language processing], and a means for [generating an answer based on the search results and formatting the answer by natural language generation while taking the emotions into consideration]. This makes it possible to provide answers that take the user's emotions into consideration and specific, useful information that takes past cases into consideration.

[0881] A "communication terminal" is a device used by a user to input a question and send it to a server, and includes a personal computer, smartphone, tablet, etc.

[0882] A "server" is an information processing device that includes a central device and software that receives questions from users, analyzes them, generates answers, and sends them to terminals.

[0883] An "emotion analysis engine" is a software module that analyzes a user's emotions from text entered by the user and identifies emotional states such as stress, anxiety, and joy.

[0884] "Natural language processing (NLP)" is a technology for analyzing questions entered by users and extracting keywords and the user's intent, and includes methods of text analysis and semantic understanding.

[0885] A "regulations database" is a database that stores employee regulations and company rules and guidelines.

[0886] A "case database" is a database that stores past cases and case studies.

[0887] "Natural language generation (NLG)" is a technology that generates sentences in a form that is easy for humans to understand based on search results and analytical information.

[0888] "Formatting with consideration for emotions" means creating a response format that is in line with the user's feelings by taking into account the user's emotional state and adding appropriate phrases to the response.

[0889] The present invention relates to an AI system that provides quick and appropriate answers to questions entered by a user through a terminal, and is characterized in that it generates answers by analyzing the user's emotions. The following describes in detail the embodiments of the present invention.

[0890] Users access the system using a communication device such as a PC or smartphone. A browser or dedicated application is installed on the communication device, and users can input and send questions. For example, suppose a user inputs a question such as, "Please tell me how to register attendance when visiting the hospital."

[0891] The device sends the questions entered by the user to the server using the HTTPS protocol, using SSL / TLS encryption technology to ensure a secure communication environment.

[0892] The server analyzes the questions received from the user. First, it uses a sentiment analysis engine to analyze the user's emotions. This sentiment analysis engine uses sentiment analysis APIs such as Azure Cognitive Services and IBM Watson. For example, if "anxiety" is detected, that information is used for subsequent processing.

[0893] The server then analyzes the question using a natural language processing (NLP) module, which utilizes technologies such as spaCy and TensorFlow. Specifically, it extracts keywords such as "hospital visit" and "attendance registration" from the question, as well as the user's intent.

[0894] The server then searches for relevant information from the policy and case databases based on the extracted keywords and intent. These databases are managed using database management systems such as MySQL and PostgreSQL. The search results include relevant information based on employee policy and past cases.

[0895] The server evaluates the search results and generates the most appropriate answer. This answer generation uses a natural language generation (NLG) module, such as OpenAI's GPT-3 or Google's BERT. At this stage, the results of the sentiment analysis engine are reflected, and the answer is formatted to take the user's emotions into consideration. For example, if the user is detected as "anxious," the answer may include phrases such as "Don't worry" or "Don't worry."

[0896] Finally, the server sends the generated answer to the device and displays it to the user. The user can check the specific answer on the screen of their communication device. For example, if the question is, "What is the procedure for taking paid leave?" and the emotion analysis engine detects "stress," the answer might include something like, "To take paid leave, please get your supervisor's approval and then apply through the attendance management system. It's okay, your health is important, so please get plenty of rest."

[0897] In this way, the system of the present invention can provide users with prompt and appropriate answers that take into consideration the user's feelings, and can provide users with accurate information based on employee regulations and past cases, which can be expected to deepen understanding of employee questions and improve work efficiency.

[0898] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0899] Step 1:

[0900] The user enters and sends a question from a communication device. The user accesses a specific web application or mobile application using a PC or smartphone and enters a question. For example, the user might enter, "Please tell me the procedure for taking paid leave." The entered question is sent to the server using the HTTPS protocol.

[0901] Input: User-entered question: "What is the procedure for taking paid leave?"

[0902] Output: The question data sent to the server

[0903] Step 2:

[0904] The device sends a query to the server. The device securely sends data to the server using SSL / TLS. The sent data is received by the server.

[0905] Input: Question data sent from the terminal

[0906] Output: The query data received by the server

[0907] Step 3:

[0908] The server analyzes the user's emotions using a sentiment analysis engine. The server inputs the received question data into the analysis engine and analyzes emotions (e.g., stress, anxiety, joy, etc.) from the question text. This analysis uses a sentiment analysis API (e.g., Azure Cognitive Services or IBM Watson).

[0909] Input: Received question data "Please tell me the procedure for taking paid leave."

[0910] Output: Emotion analysis results such as "stress"

[0911] Step 4:

[0912] The server analyzes the question using a natural language processing (NLP) module. The server passes the sentiment analysis results and question data to the NLP module, which extracts important keywords and user intent. NLP technologies (e.g., spaCy and TensorFlow) are used for this analysis.

[0913] Input: Question data "Please tell me the procedure for taking paid leave" and emotion analysis result "Stress"

[0914] Output: Keywords such as "paid leave," "acquisition," and "procedure" and user intent

[0915] Step 5:

[0916] The server searches for relevant information from the database. Based on the extracted keywords and intent, the server searches for information from the rule database and case database. At this stage, SQL queries are used to quickly obtain the required information.

[0917] Input: Keywords "paid leave", "acquisition", "procedure" and user intent

[0918] Output: Search results and past cases

[0919] Step 6:

[0920] The server generates answers and formats them taking emotions into consideration. The server inputs the search results into an NLG module, which generates answers in natural language based on the emotional information. For example, OpenAI GPT-3 and Google BERT are used in this process. Depending on the user's emotional state (e.g., stress), phrases such as "Don't worry" or "Don't worry" are added to the answer.

[0921] Input: Search results, past cases, and emotional information about "stress"

[0922] Output: Emotionally sensitive formatted answers

[0923] Step 7:

[0924] The server sends the generated response to the terminal. The server encodes the generated response in JSON format or similar and sends it to the terminal via HTTPS.

[0925] Input: Formatted answer

[0926] Output: Response data sent to the device

[0927] Step 8:

[0928] The terminal displays the answer to the user. The terminal decodes the received answer data into an appropriate format and displays it to the user. The user can check the specific answer on the terminal screen.

[0929] Input: Response data sent from the server

[0930] Output: The specific answer that is displayed to the user

[0931] The above is the specific processing flow of this system.

[0932] (Application example 2)

[0933] 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."

[0934] Conventional electronic payment systems have struggled to provide quick and appropriate answers to the various problems and questions users face. Furthermore, by returning mechanical answers without considering the user's emotional state, there is a risk of increasing the user's anxiety and stress. This can lead to a decline in user satisfaction and a deterioration in service quality. To solve this issue, a system is needed that provides answers that take the user's emotions into consideration and resolves problems quickly and accurately.

[0935] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for a user to input and send a question from a terminal, means for the server to receive the question and analyze the question using an emotion recognition engine and natural language processing, means for searching a database for rules and past cases based on the analysis results, means for generating an answer based on the search results and formatting it using natural language generation that reflects emotional information, and means for sending the generated answer to the terminal and displaying it to the user. This makes it possible to provide a quick and accurate answer while taking the user's emotional state into consideration, thereby improving user satisfaction and maintaining service quality.

[0936] The means by which "users can enter and submit questions from their devices" refers to an interface that allows users to access the system using devices such as smartphones or PCs, enter questions in text format, and submit them.

[0937] The means "the server receives the question and analyzes the question using an emotion recognition engine and natural language processing" refers to a process in which the server first analyzes the emotional state of the user's question received from the user using an emotion recognition engine, and then analyzes the content of the question using natural language processing.

[0938] The means for "searching for regulations and past cases from a database based on the analysis results" refers to a mechanism for searching for related regulation information and past cases from a database based on the analyzed question content and emotion information.

[0939] The method of "generating answers based on search results and formatting them using natural language generation that reflects emotional information" refers to the process of using the searched information to generate appropriate answers to the user's questions, and then formatting the answers in natural language using appropriate expressions, taking into account the user's emotional information.

[0940] The means for "sending the generated answer to the terminal and displaying it to the user" refers to a mechanism for sending the generated answer to the user's terminal and displaying it on the screen to provide it to the user.

[0941] The means of "selecting the most appropriate answer taking into consideration the user's emotional information" refers to the process of selecting and providing the answer from the search results that best corresponds to the user's emotional state.

[0942] The means of "using a regulations database and a database of past cases" refers to a mechanism for obtaining information from a regulations database in which organizational rules and guidelines are stored and a database in which similar past cases are stored.

[0943] The present invention relates to an AI system that recognizes a user's emotions and generates appropriate responses based on the emotions. Specific embodiments of the present invention will be described below.

[0944] First, a user accesses the system using a terminal (e.g., a smartphone or PC) and enters a question. For example, they might enter, "How do I change my credit card information?" This question is immediately sent from the terminal to the server.

[0945] The server analyzes the received question using an emotion recognition engine and a natural language processing (NLP) module. The emotion recognition engine analyzes emotions (e.g., anxiety, stress, joy, etc.) from the user's input text and obtains the results. For example, if the emotion engine detects "anxiety," that information is used in the subsequent analysis process.

[0946] The server then searches the rule database and case database based on the emotion information and the analyzed question content, thereby obtaining relevant information.

[0947] After retrieving the relevant information, the server generates an appropriate answer, taking into account the user's emotional state. A natural language generation (NLG) module assists this, formatting the answer in natural language to reflect the emotional information. For example, if the user is "feeling anxious" when asking a question, the server might generate an answer such as, "Don't worry, you can easily change your credit card information by following these steps."

[0948] Finally, the generated answer is sent to the terminal and displayed to the user, allowing the user to quickly obtain appropriate information for their question and receive an answer that takes their emotions into consideration.

[0949] Hardware and software examples

[0950] Hardware: Web server and user device (PC, smartphone)

[0951] software:

[0952] Emotion Recognition Engine: Hugging Face Transformers Library

[0953] Database: Regulation database and case database (e.g., PostgreSQL)

[0954] Natural Language Processing (NLP) module: uses OpenAI's GPT-3

[0955] Processing examples and prompts

[0956] Examples:

[0957] If a user asks "How do I change my credit card information?" and the emotion engine detects "anxiety," the server generates an answer that corresponds to the "anxious" state, such as "Don't worry, you can easily change your credit card information by following these steps."

[0958] Example prompt sentence:

[0959] Answer questions about electronic payment services when users feel uneasy. Question: How do I change my credit card information? Answer:

[0960] The present invention makes it possible to provide a prompt and appropriate response taking into account the user's emotional state, thereby improving user convenience and satisfaction.

[0961] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0962] Step 1:

[0963] The user inputs a question from the terminal and sends it.

[0964] The user enters a question using a smartphone or PC and presses the send button. For example, the user enters "How do I change my credit card information?" This input text is sent to the server.

[0965] Step 2:

[0966] The server receives the question and analyzes it using an emotion recognition engine and natural language processing.

[0967] The server processes the question received from the user. First, it uses an emotion recognition engine to analyze the sentiment of the text. In this case, "anxiety" is detected. Next, it uses a natural language processing module to analyze the keywords in the question and the user's intent. For example, the keywords "credit card information" and "change" are extracted.

[0968] Step 3:

[0969] Based on the analysis results, regulations and past cases are searched from the database.

[0970] The server then searches for relevant information from its database of company regulations and past case studies based on the analyzed keywords and intent. This operation retrieves regulations regarding changes to credit card information and similar past cases.

[0971] Step 4:

[0972] Answers are generated based on search results and formatted using natural language generation that reflects emotional information.

[0973] The server analyzes the retrieved information and generates an appropriate answer to the user's question. A natural language generation (NLG) module takes the user's sentiment into account to shape the answer. For example, the answer might be, "Don't worry, you can easily change your credit card information by following these steps."

[0974] Step 5:

[0975] The generated answer is sent to the terminal and displayed to the user.

[0976] The server generates a response and sends it to the user's device. The device receives the response and displays it to the user. The user can see the response on the device, such as "Don't worry, you can easily change your credit card information by following these steps."

[0977] 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.

[0978] 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.

[0979] 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.

[0980] [Fourth embodiment]

[0981] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0982] 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.

[0983] 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).

[0984] 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.

[0985] 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.

[0986] 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).

[0987] 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.

[0988] 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.

[0989] 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.

[0990] 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.

[0991] 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.

[0992] 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.

[0993] 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."

[0994] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases. The detailed implementation method for users to input questions and receive appropriate answers is described below.

[0995] First, the user accesses the system using a device such as a PC or smartphone and enters their question. For example, they might enter, "Please tell me how to register attendance when I go to the hospital." This question is immediately sent from the device to the server.

[0996] The server uses a natural language processing (NLP) module to analyze the questions received from the device. The server analyzes the questions and extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. This makes it clear which regulations should be referenced.

[0997] The server then searches the employee regulations database and past case database for related information based on the extracted keywords and intent, and retrieves the relevant employee regulations and similar past cases.

[0998] The server evaluates the search results and generates the most appropriate answer. The generated answer is formatted using a natural language generation (NLG) module to make it easy for the user to understand. For example, the server might generate an answer such as, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[0999] Finally, the server sends the formatted answer to the terminal, and the terminal displays the generated answer to the user, allowing the user to quickly obtain appropriate information for the question, deepening their understanding of employee regulations and improving work efficiency.

[1000] As a concrete example, consider the case where a user asks, "What is the procedure for taking paid leave?" In this case, the server extracts keywords such as "paid leave," "taking," and "procedure," and searches for relevant regulations in the employee regulations database. It then takes into account similar past cases and generates an answer showing the appropriate procedure. For example, an answer including specific steps, such as "To take paid leave, please obtain approval from your supervisor and then apply through the attendance management system," is generated and presented to the user via their device.

[1001] As described above, the system of the present invention allows employees to get quick and accurate answers to their questions. This system significantly reduces the time and effort required to check employee regulations, thereby improving work efficiency.

[1002] The processing flow will be explained below.

[1003] Step 1:

[1004] User: Accesses the system and inputs a question into the interface on the terminal. For example, inputs "How do I register my attendance when I go to the hospital?"

[1005] Terminal: Receives user input and sends questions to the server.

[1006] Step 2:

[1007] Server: Receives questions sent as HTTP requests from the terminal.

[1008] Server: Passes the received question to a natural language processing (NLP) module to analyze the text.

[1009] Specific operation: Divide the question into tokens and extract keywords (e.g., "visiting the hospital" or "attendance registration") and intent.

[1010] Step 3:

[1011] Server: Based on the analysis results, a query is generated to search the employee regulations database and the past case database.

[1012] Specific operation: Generates SQL queries, etc., and retrieves information related to the question from the database.

[1013] Step 4:

[1014] Server: Aggregates information retrieved from the database and evaluates search results.

[1015] Specific operation: The system sorts the data and selects the most relevant information, such as "A doctor's note is required when visiting the hospital."

[1016] Step 5:

[1017] Server: Generates an answer to the user's question based on the evaluation results.

[1018] How it works: Responses are constructed based on employee policies and past experience, and formatted in a way that is easy for humans to understand. A natural language generation (NLG) module assists in this process.

[1019] Example: "When visiting a hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[1020] Step 6:

[1021] Server: Sends the generated answer to the device.

[1022] Specific operation: The formatted response is returned to the terminal in JSON format or similar.

[1023] Step 7:

[1024] Terminal: Displays the answer received from the server on a user interface.

[1025] Specific behavior: Updates the UI to present the received information to the user in the appropriate format.

[1026] Step 8:

[1027] User: Check the answers displayed on the device and take action if necessary.

[1028] Example: Take specific action such as registering as "visiting the hospital" in the attendance management system.

[1029] Through these steps, users can get instant and accurate answers to their questions, streamlining the employee policy review process and improving work productivity.

[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 robot 414 will be referred to as a "terminal."

[1032] In modern companies, having employees be able to quickly and accurately refer to employee regulations and past cases directly leads to improved work efficiency. However, current systems require employees to manually search for regulations and cases, which is time-consuming and labor-intensive. To solve this problem, there is a need for a system that allows users to obtain appropriate answers simply by entering a question.

[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 a means for a user to input and send a query from a computer terminal, a means for the server to receive the query and analyze it using natural language processing, and a means for searching for regulations and past cases from a database based on the analysis results, thereby enabling a user to quickly and accurately refer to employee regulations and past cases.

[1035] A "user" is a person who operates a computer terminal to input a query and obtain a response from the system.

[1036] "Computer terminal" refers to a device that can connect to the Internet, such as a PC or smartphone.

[1037] A "query" is a question or command that a user enters into a system with the purpose of obtaining information.

[1038] A "server" is a computer system that receives inquiries, performs natural language processing and search processing, and generates and transmits answers.

[1039] "Natural language processing" is a technique used by computers to analyze human language and understand user queries.

[1040] "Analysis" is the process by which the server understands the content of the query and extracts relevant information.

[1041] A "database" is a collection of information in which information such as regulations and past cases is structured and stored.

[1042] "Policies" are documents or information that describe company rules and procedures that employees must follow.

[1043] "Past cases" are records of responses and actions taken in response to similar inquiries that have occurred in the past.

[1044] "Searching" is the process of finding specific information within a database.

[1045] A "response" is information or instructions that the server generates and sends in response to a user's inquiry.

[1046] "Generation" is the process by which the server creates a response for the user based on the analysis results.

[1047] "Natural language generation" is a technology that allows a computer to create text in a form that humans can understand.

[1048] "Transmission" means the act of electronically transferring a query or response from one party to another.

[1049] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases. Specific embodiments are described below.

[1050] Hardware and software used

[1051] This system mainly consists of the following hardware and software:

[1052] Hardware

[1053] 1. Computer terminal

[1054] A PC or smartphone for users to enter questions

[1055] 2. Server

[1056] A high-performance computer for analyzing questions, searching databases, and generating answers

[1057] software

[1058] 1. Natural Language Processing (NLP) Module

[1059] Software used to analyze user questions. Examples include NLP libraries such as spaCy and Transformers.

[1060] 2. Database Management System (DBMS)

[1061] It is used to execute SQL queries to search for employee policies and past cases. Examples include MySQL and PostgreSQL.

[1062] 3. Natural Language Generation (NLG) Module

[1063] It is used to generate answers based on search results and format them in a format that is easy for users to understand. Specific examples include OpenAI's GPT-3 and GPT-4.

[1064] Data processing and data calculation

[1065] 1. Question analysis

[1066] The question entered by the user on the computer terminal is sent to the server, which uses an NLP module to extract keywords such as "hospital visit" and "attendance registration" from the question and the user's intention.

[1067] 2. Database search

[1068] The server searches for relevant information from the employee regulations database and past case database based on the extracted keywords. This search uses SQL queries. For example, the SQL query for the keyword "hospital visit" is executed as follows: "SELECT FROM employee_regulations WHERE keyword='hospital visit'".

[1069] 3. Answer generation

[1070] The server evaluates the search results and generates the most appropriate answer using an NLG module, which is formatted in a way that is easy for the user to understand, such as "To take paid leave, please get your supervisor's approval and apply through the attendance management system."

[1071] Specific examples

[1072] A specific example will be given below in which a user asks on a computer terminal, "Please tell me how to register attendance when visiting a hospital."

[1073] 1. Enter your question

[1074] Users access the system using a PC or smartphone and enter their questions.

[1075] 2. Submit a question

[1076] The terminal immediately sends the query to the server.

[1077] 3. Question analysis

[1078] The server uses an NLP module to analyze the question and extract the keywords "hospital visit" and "attendance registration."

[1079] 4. Database Search

[1080] The server searches the employee regulations database and past case database for relevant information and acquires the necessary information.

[1081] 5. Answer generation

[1082] Based on the search results, the server uses the NLG module to generate a response such as, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system."

[1083] 6. Submit your response

[1084] The server sends the generated answer to the computer terminal, which displays it to the user.

[1085] Prompt Sentence Examples

[1086] "Please tell me how to register attendance when visiting the hospital."

[1087] What is the procedure for taking paid leave?

[1088] "I would like to know how to claim expenses when working from home."

[1089] By entering the above prompts, the user can quickly and accurately obtain information on employee regulations and past cases. The system of the present invention provides prompt and appropriate answers to the user's questions, significantly improving the work efficiency of employees.

[1090] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1091] Step 1:

[1092] A user accesses the system using a computer terminal and inputs a question. Specifically, the user opens a browser, accesses the system's web interface, and inputs a question into the input form. For example, the user might input, "Please tell me how to register attendance when visiting the hospital." This question becomes input data.

[1093] Step 2:

[1094] The terminal immediately sends the question entered by the user to the server. Specifically, when the user clicks the send button, the terminal sends the input data to the server as an HTTP POST request. This transmission becomes the input for processing.

[1095] Step 3:

[1096] The server analyzes the questions received from the device using a natural language processing (NLP) module. Specifically, the server tokenizes the questions received using an NLP library (e.g., spaCy or Transformers) and performs semantic analysis. This analysis extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. The analysis results are the output for data processing.

[1097] Step 4:

[1098] The server searches for relevant information from the employee regulations database and past case database based on the analysis results. Specifically, it generates an SQL query based on the extracted keywords and searches for data using a database management system such as MySQL or PostgreSQL. For example, an SQL query such as "SELECT FROM employee_regulations WHERE keyword='hospital visit'" is executed. The data retrieved by this search becomes the output of data calculations based on the input data.

[1099] Step 5:

[1100] The server evaluates the search results and generates the most appropriate answer. Specifically, it uses an NLG module (for example, OpenAI's GPT-3) to generate natural language based on the search results, and creates an answer in a format that is easy for the user to understand. For example, the answer could be something like, "When visiting the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system." This generated answer becomes the output data.

[1101] Step 6:

[1102] The server sends the generated answer to the terminal. Specifically, it organizes the generated answer in JSON format and sends it to the terminal as an HTTP response. This transmission becomes the output of the processing.

[1103] Step 7:

[1104] The device displays the answer received from the server to the user. Specifically, it updates the DOM of the web page and reflects the received text in the answer display area. As a result, the user can see the appropriate information for the question on the screen. This display is the final output.

[1105] (Application example 1)

[1106] 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."

[1107] In brick-and-mortar stores, when store clerks or managers have questions about store operations, it can be difficult to quickly and accurately obtain the appropriate information. For example, when they need to immediately check product return policies or complaint handling procedures, not being able to obtain the appropriate information in a timely manner can reduce work efficiency. Another issue is the lack of a way to easily check information on a device at hand while working on other tasks.

[1108] 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.

[1109] In this invention, the server includes means for a user to input and send a question from a terminal, means for the server to receive the question and analyze the question using natural language processing, means for searching a database for regulations and past cases based on the analysis results, means for generating an answer based on the search results and formatting it using natural language generation, and means for transmitting the generated answer to the terminal and displaying it to the user, thereby enabling users to quickly and accurately obtain appropriate information for questions about store operation in a physical store.

[1110] "Means for users to input and send questions from a terminal" refers to the process and interface by which a user inputs a specific question using a smart device and sends the question to a server over a network.

[1111] "Means for the server to receive a question and analyze the question using natural language processing" refers to a module or algorithm that allows the server to receive a question sent by a user, analyze the question using natural language processing technology, and understand its meaning and intent.

[1112] "Means for searching for regulations and past cases from a database based on the analysis results" refers to functions and modules that extract keywords from the results of the analyzed questions and use them to search a database in which regulations and past cases are stored.

[1113] "Means of generating answers based on search results and formatting them using natural language generation" refers to the function of using natural language generation technology to create answers in a format that is easy for users to understand based on information obtained from a database, and then formatting them.

[1114] "Means for transmitting the generated answer to the terminal and displaying it to the user" refers to the processes and techniques for transmitting the formatted answer back over the network to the originating smart device and visually displaying it to the user.

[1115] "A means of providing work support by displaying generated answers on a smart device" refers to a function that displays generated answers on smart devices such as smartphones, tablets, and smart glasses, and supports work related to store operations.

[1116] "Means of utilizing regulations related to store operations and databases of past cases" refers to the ability to utilize databases that record regulations related to store operations (such as refund policies and complaint handling procedures) and specific past cases.

[1117] This invention provides a question-answering system that uses AI technology to improve the efficiency of brick-and-mortar store operations. Specific embodiments of this system are described below.

[1118] First, a user (a store clerk or manager) uses a smart device (smartphone, tablet, smart glasses, etc.) to input and send a question. For example, they might input, "Please tell me the procedure for returning a product." This question is immediately sent to the server via the network.

[1119] The server uses a natural language processing (NLP) module such as Spacy to analyze the received question. Specifically, it extracts keywords such as "return" and "procedure" from the question and understands the user's intent.

[1120] The server then searches a database for relevant regulations and past cases based on the extracted keywords. The regulation database includes product return policies and store operation rules, while the past case database stores specific past cases. This allows for quick retrieval of the relevant information.

[1121] The server then generates an answer based on the information it has obtained, using natural language generation (NLG) modules like Hugging Face Transformers to format the answer in a way that is easy for the user to understand, such as specific instructions like "When returning an item, please bring the item with the receipt and fill out the return form."

[1122] The generated answer is sent to the original smart device via the network, where the user can visually check it and take any necessary action immediately, significantly improving the efficiency of store operations.

[1123] For example:

[1124] When a user asks, "What is the procedure for returning a product?", the server extracts keywords such as "return" and "procedure" and searches the database for relevant regulations and past cases. It then generates a response such as, "When returning a product, please bring the product along with the receipt and fill out the necessary information on the return form," which is displayed on the smart device. This process allows the user to take prompt and appropriate action.

[1125] Example prompt sentence:

[1126] Q: What is the procedure for returning a product?

[1127] A: When returning an item, please bring the item with you along with your receipt and fill out the return form.

[1128] This system makes it possible to quickly resolve questions about store operations and improve operational efficiency.

[1129] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1130] Step 1:

[1131] A user uses a smart device (smartphone, tablet, smart glasses, etc.) to input a question and press the "Send" button. For example, the user might input "Please tell me the procedure for returning a product." At this time, the input question is sent as text data from the device to the server.

[1132] Input: The question text entered by the user (e.g., "How do I return an item?")

[1133] Output: Submitted question text

[1134] Step 2:

[1135] The server receives the question sent from the device and analyzes it using a natural language processing (NLP) module that utilizes Spacy. Specifically, the question is tokenized and keywords such as "return" and "procedure" are extracted. The user's intent is understood based on the extracted keywords.

[1136] Input: Submitted question text

[1137] Output: Extracted keywords and user intent (e.g., "return," "procedure")

[1138] Step 3:

[1139] The server searches the regulatory database and past case database based on the extracted keywords, thereby obtaining relevant regulatory information on store operations and past case studies that have previously addressed similar questions. The search results are stored in list format.

[1140] Input: Extracted keywords and user intent

[1141] Output: Search result list (e.g. related regulatory information and past cases)

[1142] Step 4:

[1143] Based on the search results, the server uses a natural language generation (NLG) module that utilizes Hugging Face Transformers to generate answers in a format that is easy for the user to understand, specifically outputting answers in natural-sounding sentences while taking into account the context of the search results.

[1144] Input: Search result list

[1145] Output: The generated answer (e.g., "When returning an item, please bring the item with you along with your receipt and fill out the return form.")

[1146] Step 5:

[1147] The server sends the generated answer to the original smart device via the network, and the smart device displays the received answer on its interface so that the user can visually confirm it.

[1148] Input: Generated Answer

[1149] Output: Answer displayed on the user's terminal

[1150] Step 6:

[1151] The user checks the answers displayed on the smart device and takes specific action, such as actually returning the product based on the answers.

[1152] Input: Answer displayed on smart device

[1153] Output: User's actual action (e.g., returning a product)

[1154] In this way, by responding quickly and accurately to questions about running a physical store, the operational efficiency of the entire store can be improved.

[1155] 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.

[1156] The present invention relates to an AI system that quickly and accurately provides employee regulations and past cases, and by combining it with an emotion engine that recognizes the user's emotions, it is possible to obtain more appropriate answers. Specific embodiments of the present invention are described below.

[1157] First, the user accesses the system using a device such as a PC or smartphone and enters a question. For example, they might enter, "Please tell me how to register attendance when I go to the hospital." This question is immediately sent from the device to the server.

[1158] The server first uses an emotion engine to analyze the question received from the terminal. The emotion engine analyzes emotions (e.g., stress, anxiety, anger, joy, etc.) from the user's input text and obtains the results. For example, if the emotion engine detects "anxiety" from the user's question, that information is used in the subsequent analysis process.

[1159] The server then passes the acquired emotional information along with the question to a natural language processing (NLP) module. The server analyzes the question and extracts keywords such as "hospital visit" and "attendance registration" as well as the user's intent. This makes it clear which regulations should be referenced.

[1160] The server searches the employee regulations database and past case database for related information based on the extracted keywords and intent, and obtains the relevant employee regulations and similar past cases.

[1161] The server evaluates the search results and generates the most appropriate answer. The generated answer is adjusted to reflect the user's emotional information. For example, if the user is feeling anxious when asking a question, the answer will include phrases such as "Don't worry" or "Don't worry," providing an answer that takes the user's emotions into consideration. This is supported by a natural language generation (NLG) module.

[1162] Finally, the server sends the formatted answer to the terminal, and the terminal displays the generated answer to the user. This allows the user to quickly obtain appropriate information for the question and receive an answer that takes into consideration their feelings, thereby deepening their understanding of employee regulations and improving work efficiency.

[1163] As a concrete example, consider the case where a user asks, "What is the procedure for taking paid leave?" and the emotion engine detects "stress." The server extracts keywords such as "paid leave," "taking," and "procedure," and searches for relevant regulations in the employee regulations database. It then generates an answer showing the appropriate procedure, taking into account similar past cases. For example, an answer that takes specific steps and emotions into consideration, such as "To take paid leave, please get your supervisor's approval and then apply through the attendance management system. It's okay, your health is important, so please take plenty of rest," is generated and presented to the user via their device.

[1164] As described above, by recognizing the user's emotions and adjusting responses accordingly, the system of the present invention not only allows employees to get quick and accurate answers to their questions, but also allows for more user-friendly responses by taking emotions into consideration, thereby streamlining the confirmation process for employee regulations and improving work productivity.

[1165] The processing flow will be explained below.

[1166] Step 1:

[1167] User: Accesses the system using a device such as a PC or smartphone and enters a question. For example, "Please tell me how to register attendance when going to the hospital."

[1168] Terminal: Receives user input and sends questions to the server.

[1169] Step 2:

[1170] Server: Receives questions sent as HTTP requests from the terminal.

[1171] Server: Passes the received question to the emotion engine and analyzes the user's emotion from the text.

[1172] Specific operation: Analyzes the question and applies an emotion model to extract emotions such as "stress," "anxiety," and "joy." For example, the emotion "anxiety" can be detected from the question "Please tell me how to register attendance when going to the hospital."

[1173] Step 3:

[1174] Server: Holds the extracted sentiment information and passes the question to a natural language processing (NLP) module to analyze the text.

[1175] Specific operation: The question is divided into tokens, and keywords (e.g., "visiting the hospital" or "attendance registration") and the user's intent are extracted.

[1176] Step 4:

[1177] Server: Based on the analysis results, a query is generated to search the employee regulations database and the past case database.

[1178] Specific operation: Generates SQL queries, etc., and retrieves information related to the question from the database.

[1179] Step 5:

[1180] Server: Aggregates information retrieved from the database and evaluates search results.

[1181] Specific behavior: Orders the retrieved data and identifies the most relevant information, such as "A doctor's note is required when visiting the hospital."

[1182] Step 6:

[1183] Server: Based on the evaluation results, it generates an answer to the user's question. It adjusts the answer taking into account emotional information.

[1184] How it works: The server constructs a response based on employee policy and past case data, and a natural language generation (NLG) module formats it. For example, if anxiety is detected, a phrase like "Don't worry" is added to the response.

[1185] Example: "When you visit the hospital, please attach a doctor's certificate and register it as a 'hospital visit' in the attendance management system. Don't worry, everything will be fine."

[1186] Step 7:

[1187] Server: Sends the generated answer to the device.

[1188] Specific operation: The formatted response is returned to the terminal in JSON format or similar.

[1189] Step 8:

[1190] Terminal: Displays the answer received from the server on a user interface.

[1191] Specific behavior: Updates the UI to present the received information to the user in the appropriate format.

[1192] Step 9:

[1193] User: Check the answers displayed on the device and take action if necessary.

[1194] Example: Take specific action such as registering as "visiting the hospital" in the attendance management system.

[1195] This allows users to receive instant, accurate answers to their questions, and also allows them to receive answers that take into consideration their feelings, which improves the efficiency of employee policy confirmation work and increases work productivity.

[1196] Example 2

[1197] 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."

[1198] Conventional information provision systems have had the problem that answers to user questions do not take emotions into consideration, resulting in low user satisfaction. Another problem is that it is difficult to accurately understand the content of the question entered by the user and provide a quick and appropriate answer. Furthermore, there is a need for systems that can provide more specific and useful information by generating answers that take past cases into account.

[1199] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for [analyzing the user's emotions using an emotion analysis engine], a means for [analyzing the question by natural language processing], and a means for [generating an answer based on the search results and formatting the answer by natural language generation while taking the emotions into consideration]. This makes it possible to provide answers that take the user's emotions into consideration and specific, useful information that takes past cases into consideration.

[1200] A "communication terminal" is a device used by a user to input a question and send it to a server, and includes a personal computer, smartphone, tablet, etc.

[1201] A "server" is an information processing device that includes a central device and software that receives questions from users, analyzes them, generates answers, and sends them to terminals.

[1202] An "emotion analysis engine" is a software module that analyzes a user's emotions from text entered by the user and identifies emotional states such as stress, anxiety, and joy.

[1203] "Natural language processing (NLP)" is a technology for analyzing questions entered by users and extracting keywords and the user's intent, and includes methods of text analysis and semantic understanding.

[1204] A "regulations database" is a database that stores employee regulations and company rules and guidelines.

[1205] A "case database" is a database that stores past cases and case studies.

[1206] "Natural language generation (NLG)" is a technology that generates sentences in a form that is easy for humans to understand based on search results and analytical information.

[1207] "Formatting with consideration for emotions" means creating a response format that is in line with the user's feelings by taking into account the user's emotional state and adding appropriate phrases to the response.

[1208] The present invention relates to an AI system that provides quick and appropriate answers to questions entered by a user through a terminal, and is characterized in that it generates answers by analyzing the user's emotions. The following describes in detail the embodiments of the present invention.

[1209] Users access the system using a communication device such as a PC or smartphone. A browser or dedicated application is installed on the communication device, and users can input and send questions. For example, suppose a user inputs a question such as, "Please tell me how to register attendance when visiting the hospital."

[1210] The device sends the questions entered by the user to the server using the HTTPS protocol, using SSL / TLS encryption technology to ensure a secure communication environment.

[1211] The server analyzes the questions received from the user. First, it uses a sentiment analysis engine to analyze the user's emotions. This sentiment analysis engine uses sentiment analysis APIs such as Azure Cognitive Services and IBM Watson. For example, if "anxiety" is detected, that information is used for subsequent processing.

[1212] The server then analyzes the question using a natural language processing (NLP) module, which utilizes technologies such as spaCy and TensorFlow. Specifically, it extracts keywords such as "hospital visit" and "attendance registration" from the question, as well as the user's intent.

[1213] The server then searches for relevant information from the policy and case databases based on the extracted keywords and intent. These databases are managed using database management systems such as MySQL and PostgreSQL. The search results include relevant information based on employee policy and past cases.

[1214] The server evaluates the search results and generates the most appropriate answer. This answer generation uses a natural language generation (NLG) module, such as OpenAI's GPT-3 or Google's BERT. At this stage, the results of the sentiment analysis engine are reflected, and the answer is formatted to take the user's emotions into consideration. For example, if the user is detected as "anxious," the answer may include phrases such as "Don't worry" or "Don't worry."

[1215] Finally, the server sends the generated answer to the device and displays it to the user. The user can check the specific answer on the screen of their communication device. For example, if the question is, "What is the procedure for taking paid leave?" and the emotion analysis engine detects "stress," the answer might include something like, "To take paid leave, please get your supervisor's approval and then apply through the attendance management system. It's okay, your health is important, so please get plenty of rest."

[1216] In this way, the system of the present invention can provide users with prompt and appropriate answers that take into consideration the user's feelings, and can provide users with accurate information based on employee regulations and past cases, which can be expected to deepen understanding of employee questions and improve work efficiency.

[1217] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1218] Step 1:

[1219] The user enters and sends a question from a communication device. The user accesses a specific web application or mobile application using a PC or smartphone and enters a question. For example, the user might enter, "Please tell me the procedure for taking paid leave." The entered question is sent to the server using the HTTPS protocol.

[1220] Input: User-entered question: "What is the procedure for taking paid leave?"

[1221] Output: The question data sent to the server

[1222] Step 2:

[1223] The device sends a query to the server. The device securely sends data to the server using SSL / TLS. The sent data is received by the server.

[1224] Input: Question data sent from the terminal

[1225] Output: The query data received by the server

[1226] Step 3:

[1227] The server analyzes the user's emotions using a sentiment analysis engine. The server inputs the received question data into the analysis engine and analyzes emotions (e.g., stress, anxiety, joy, etc.) from the question text. This analysis uses a sentiment analysis API (e.g., Azure Cognitive Services or IBM Watson).

[1228] Input: Received question data "Please tell me the procedure for taking paid leave."

[1229] Output: Emotion analysis results such as "stress"

[1230] Step 4:

[1231] The server analyzes the question using a natural language processing (NLP) module. The server passes the sentiment analysis results and question data to the NLP module, which extracts important keywords and user intent. NLP technologies (e.g., spaCy and TensorFlow) are used for this analysis.

[1232] Input: Question data "Please tell me the procedure for taking paid leave" and emotion analysis result "Stress"

[1233] Output: Keywords such as "paid leave," "acquisition," and "procedure" and user intent

[1234] Step 5:

[1235] The server searches for relevant information from the database. Based on the extracted keywords and intent, the server searches for information from the rule database and case database. At this stage, SQL queries are used to quickly obtain the required information.

[1236] Input: Keywords "paid leave", "acquisition", "procedure" and user intent

[1237] Output: Search results and past cases

[1238] Step 6:

[1239] The server generates answers and formats them taking emotions into consideration. The server inputs the search results into an NLG module, which generates answers in natural language based on the emotional information. For example, OpenAI GPT-3 and Google BERT are used in this process. Depending on the user's emotional state (e.g., stress), phrases such as "Don't worry" or "Don't worry" are added to the answer.

[1240] Input: Search results, past cases, and emotional information about "stress"

[1241] Output: Emotionally sensitive formatted answers

[1242] Step 7:

[1243] The server sends the generated response to the terminal. The server encodes the generated response in JSON format or similar and sends it to the terminal via HTTPS.

[1244] Input: Formatted answer

[1245] Output: Response data sent to the device

[1246] Step 8:

[1247] The terminal displays the answer to the user. The terminal decodes the received answer data into an appropriate format and displays it to the user. The user can check the specific answer on the terminal screen.

[1248] Input: Response data sent from the server

[1249] Output: The specific answer that is displayed to the user

[1250] The above is the specific processing flow of this system.

[1251] (Application example 2)

[1252] 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."

[1253] Conventional electronic payment systems have struggled to provide quick and appropriate answers to the various problems and questions users face. Furthermore, by returning mechanical answers without considering the user's emotional state, there is a risk of increasing the user's anxiety and stress. This can lead to a decline in user satisfaction and a deterioration in service quality. To solve this issue, a system is needed that provides answers that take the user's emotions into consideration and resolves problems quickly and accurately.

[1254] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for a user to input and send a question from a terminal, means for the server to receive the question and analyze the question using an emotion recognition engine and natural language processing, means for searching a database for rules and past cases based on the analysis results, means for generating an answer based on the search results and formatting it using natural language generation that reflects emotional information, and means for sending the generated answer to the terminal and displaying it to the user. This makes it possible to provide a quick and accurate answer while taking the user's emotional state into consideration, thereby improving user satisfaction and maintaining service quality.

[1255] The means by which "users can enter and submit questions from their devices" refers to an interface that allows users to access the system using devices such as smartphones or PCs, enter questions in text format, and submit them.

[1256] The means "the server receives the question and analyzes the question using an emotion recognition engine and natural language processing" refers to a process in which the server first analyzes the emotional state of the user's question received from the user using an emotion recognition engine, and then analyzes the content of the question using natural language processing.

[1257] The means for "searching for regulations and past cases from a database based on the analysis results" refers to a mechanism for searching for related regulation information and past cases from a database based on the analyzed question content and emotion information.

[1258] The method of "generating answers based on search results and formatting them using natural language generation that reflects emotional information" refers to the process of using the searched information to generate appropriate answers to the user's questions, and then formatting the answers in natural language using appropriate expressions, taking into account the user's emotional information.

[1259] The means for "sending the generated answer to the terminal and displaying it to the user" refers to a mechanism for sending the generated answer to the user's terminal and displaying it on the screen to provide it to the user.

[1260] The means of "selecting the most appropriate answer taking into consideration the user's emotional information" refers to the process of selecting and providing the answer from the search results that best corresponds to the user's emotional state.

[1261] The means of "using a regulations database and a database of past cases" refers to a mechanism for obtaining information from a regulations database in which organizational rules and guidelines are stored and a database in which similar past cases are stored.

[1262] The present invention relates to an AI system that recognizes a user's emotions and generates appropriate responses based on the emotions. Specific embodiments of the present invention will be described below.

[1263] First, a user accesses the system using a terminal (e.g., a smartphone or PC) and enters a question. For example, they might enter, "How do I change my credit card information?" This question is immediately sent from the terminal to the server.

[1264] The server analyzes the received question using an emotion recognition engine and a natural language processing (NLP) module. The emotion recognition engine analyzes emotions (e.g., anxiety, stress, joy, etc.) from the user's input text and obtains the results. For example, if the emotion engine detects "anxiety," that information is used in the subsequent analysis process.

[1265] The server then searches the rule database and case database based on the emotion information and the analyzed question content, thereby obtaining relevant information.

[1266] After retrieving the relevant information, the server generates an appropriate answer, taking into account the user's emotional state. A natural language generation (NLG) module assists this, formatting the answer in natural language to reflect the emotional information. For example, if the user is "feeling anxious" when asking a question, the server might generate an answer such as, "Don't worry, you can easily change your credit card information by following these steps."

[1267] Finally, the generated answer is sent to the terminal and displayed to the user, allowing the user to quickly obtain appropriate information for their question and receive an answer that takes their emotions into consideration.

[1268] Hardware and software examples

[1269] Hardware: Web server and user device (PC, smartphone)

[1270] software:

[1271] Emotion Recognition Engine: Hugging Face Transformers Library

[1272] Database: Regulation database and case database (e.g., PostgreSQL)

[1273] Natural Language Processing (NLP) module: uses OpenAI's GPT-3

[1274] Processing examples and prompts

[1275] Examples:

[1276] If a user asks "How do I change my credit card information?" and the emotion engine detects "anxiety," the server generates an answer that corresponds to the "anxious" state, such as "Don't worry, you can easily change your credit card information by following these steps."

[1277] Example prompt sentence:

[1278] Answer questions about electronic payment services when users feel uneasy. Question: How do I change my credit card information? Answer:

[1279] The present invention makes it possible to provide a prompt and appropriate response taking into account the user's emotional state, thereby improving user convenience and satisfaction.

[1280] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1281] Step 1:

[1282] The user inputs a question from the terminal and sends it.

[1283] The user enters a question using a smartphone or PC and presses the send button. For example, the user enters "How do I change my credit card information?" This input text is sent to the server.

[1284] Step 2:

[1285] The server receives the question and analyzes it using an emotion recognition engine and natural language processing.

[1286] The server processes the question received from the user. First, it uses an emotion recognition engine to analyze the sentiment of the text. In this case, "anxiety" is detected. Next, it uses a natural language processing module to analyze the keywords in the question and the user's intent. For example, the keywords "credit card information" and "change" are extracted.

[1287] Step 3:

[1288] Based on the analysis results, regulations and past cases are searched from the database.

[1289] The server then searches for relevant information from its database of company regulations and past case studies based on the analyzed keywords and intent. This operation retrieves regulations regarding changes to credit card information and similar past cases.

[1290] Step 4:

[1291] Answers are generated based on search results and formatted using natural language generation that reflects emotional information.

[1292] The server analyzes the retrieved information and generates an appropriate answer to the user's question. A natural language generation (NLG) module takes the user's sentiment into account to shape the answer. For example, the answer might be, "Don't worry, you can easily change your credit card information by following these steps."

[1293] Step 5:

[1294] The generated answer is sent to the terminal and displayed to the user.

[1295] The server generates a response and sends it to the user's device. The device receives the response and displays it to the user. The user can see the response on the device, such as "Don't worry, you can easily change your credit card information by following these steps."

[1296] 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.

[1297] 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.

[1298] 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.

[1299] 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.

[1300] 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.

[1301] 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.

[1302] 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).

[1303] 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.

[1304] 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."

[1305] 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.

[1306] 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).

[1307] 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.

[1308] 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.

[1309] 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.

[1310] 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.

[1311] 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.

[1312] 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.

[1313] 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.

[1314] 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.

[1315] 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.

[1316] 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.

[1317] The following is further disclosed regarding the above embodiment.

[1318] (Claim 1)

[1319] A means for users to enter and submit questions from their devices;

[1320] a means for the server to receive the query and analyze the query using natural language processing;

[1321] [Searching for employee regulations and past cases from the database based on the analysis results];

[1322] A means to generate answers based on search results and format them using natural language generation;

[1323] a means for transmitting the generated answer to the terminal and displaying it to the user;

[1324] A system including:

[1325] (Claim 2)

[1326] 10. The system of claim 1, further comprising: means for evaluating search results and selecting a most appropriate answer.

[1327] (Claim 3)

[1328] The system of claim 1, further comprising means for [utilizing an employee policy database and a past case database].

[1329] "Example 1"

[1330] (Claim 1)

[1331] A means for a user to input and send a query from a computer terminal;

[1332] a means for the server to receive the query and analyze the query using natural language processing;

[1333] A means to search for regulations and past cases from the database based on the analysis results;

[1334] A means to generate answers based on search results and format them using natural language generation;

[1335] means for transmitting the generated answer to a computer terminal and displaying it to the user;

[1336] A system including:

[1337] (Claim 2)

[1338] The system of claim 1 [evaluates search results and selects the most appropriate answer].

[1339] (Claim 3)

[1340] The system described in claim 1 [utilizing a regulations database and a past case database].

[1341] "Application Example 1"

[1342] (Claim 1)

[1343] A means for users to enter and submit questions from their devices;

[1344] a means for the server to receive the query and analyze the query using natural language processing;

[1345] A means to search for regulations and past cases from the database based on the analysis results;

[1346] A means to generate answers based on search results and format them using natural language generation;

[1347] a means for transmitting the generated answer to the terminal and displaying it to the user;

[1348] A system including:

[1349] (Claim 2)

[1350] The system of claim 1, further comprising: means for providing work support by displaying the generated answers on a smart device.

[1351] (Claim 3)

[1352] The system of claim 1, further comprising: means for utilizing a database of regulations and past cases related to store operations.

[1353] "Example 2: Combining Emotion Engines"

[1354] (Claim 1)

[1355] A means for a user to input and send a question from a communication terminal;

[1356] A server receives the question and analyzes the user's emotions using a sentiment analysis engine;

[1357] a means for the server to analyze the query using natural language processing;

[1358] A means to search for regulations and past cases from the database based on the analysis results;

[1359] [Generating answers based on search results and formatting them with natural language generation to take sentiment into account],

[1360] means for transmitting the generated answer to a communication terminal and displaying the answer to a user;

[1361] A system including:

[1362] (Claim 2)

[1363] 10. The system of claim 1, further comprising: means for evaluating search results and selecting a most appropriate answer.

[1364] (Claim 3)

[1365] The system of claim 1, further comprising: means for utilizing a prescription database and a case database.

[1366] "Application example 2 when combining emotion engines"

[1367] (Claim 1)

[1368] A means for users to enter and submit questions from their devices;

[1369] A means for the server to receive the question and analyze the question using an emotion recognition engine and natural language processing;

[1370] A means to search for regulations and past cases from the database based on the analysis results;

[1371] [Generating answers based on search results and formatting them using natural language generation that reflects emotional information];

[1372] a means for transmitting the generated answer to the terminal and displaying it to the user;

[1373] A system including:

[1374] (Claim 2)

[1375] 10. The system of claim 1, further comprising: means for evaluating search results and selecting the most appropriate answer taking into account the user's emotional information.

[1376] (Claim 3)

[1377] The system of claim 1, further comprising means for [utilizing a prescribed database and a past case database]. [Explanation of symbols]

[1378] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for a user to input and send a question from a terminal; a server receiving the question and analyzing the question using natural language processing; A means for searching employee regulations and past cases from a database based on the analysis results; A means of generating answers based on search results and formatting them using natural language generation; means for transmitting the generated answer to a terminal and displaying it to a user; A system including:

2. 10. The system of claim 1, further comprising means for evaluating the search results and selecting the most appropriate answer.

3. 10. The system of claim 1, further comprising means for utilizing an employee policy database and a case history database.

Images