System

Abstract

A system is provided.SOLUTION: A system comprising: means for receiving an inquiry from a user; means for analyzing content of the received inquiry; means for generating an appropriate answer based on the content of the inquiry; means for providing the generated answer to the user; means for receiving feedback for the provided answer; and means for using the received feedback as learning data.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] In today's local governments, the aging of residents is creating a demand for attentive, individualized support, while the number of local government employees continues to decline. As a result, labor shortages and delays in responding to inquiries at government offices have become serious problems. If these issues are left unaddressed, the quality of public service may decline and public dissatisfaction may increase. Therefore, the objective of this invention is to improve the efficiency of inquiry responses and application processing at government offices, reduce the burden on employees, and improve public service. [Means for solving the problem]

[0005] According to the present invention, the above-mentioned problems are solved by the following means. Specifically, a system is provided that includes means for receiving inquiries from users, means for analyzing the received inquiries, means for generating appropriate answers based on the inquiries, means for providing the generated answers to users, means for receiving feedback on the provided answers, and means for using the received feedback as training data. This system can improve the efficiency of inquiries handled by government offices and the quality of citizen services. Furthermore, the system analyzes the contents of user inquiries using a natural language processing engine and generates answers based on the analysis results using a generative AI model. In this way, even complex inquiries can be responded to quickly and accurately, reducing the workload of employees.

[0006] "User" refers to a citizen or user of a government office who makes an inquiry using the system.

[0007] An "inquiry" is a request for information or application processing made by a user to a government office.

[0008] The "receiving means" is a function or process for incorporating the contents of a user's inquiry into the system.

[0009] "Means for analysis" refers to the function or process for analyzing the content of the received inquiry and clarifying the intent and the necessary response.

[0010] "Answer generation means" refers to the function or process for creating an appropriate answer based on the analysis results.

[0011] The "means of providing" is a function or process for communicating the generated answer to the user.

[0012] A "means for receiving feedback" is a function or process for capturing ratings and opinions on answers provided by users.

[0013] "Means for using it as learning data" refers to the function or process of using received feedback to improve the system or increase the accuracy of answers.

[0014] A "natural language processing engine" is a technology or software that analyzes text data and understands its context and meaning.

[0015] A "generative AI model" is an algorithm or software that uses artificial intelligence technology to generate appropriate answers based on analysis results. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11]FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

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

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

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

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

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

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] The present invention relates to an automatic response system that accepts inquiries from users, analyzes them, and generates and provides appropriate answers, thereby improving the efficiency of government office operations and improving public services. The following describes specific embodiments of the present invention.

[0038] Overall system configuration

[0039] The system consists of the following main components:

[0040] 1. User Device

[0041] It is a device that has an interface that allows citizens to make inquiries using government websites or applications.

[0042] 2. Server

[0043] A central computer system that receives and processes user inquiries, including analyzing the inquiry, generating answers, and receiving feedback.

[0044] 3. Natural Language Processing Engine

[0045] This is software or a program that analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to subsequent processing.

[0046] 4. Generative AI Models

[0047] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[0048] 5. Database

[0049] It is a storage system for saving data such as inquiry details, generated answers, and feedback from users.

[0050] Program processing

[0051] The program processing of this system will be explained in natural language below.

[0052] 1. Receiving Inquiries

[0053] Terminal: The user enters an inquiry about how to issue a resident registration card, etc., through the government website or application and presses the send button.

[0054] Server: Receives the inquiry sent from the terminal and temporarily stores it in a database.

[0055] 2. Analysis of inquiry content

[0056] Server: Sends the received query content to the natural language processing engine.

[0057] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident registration card," "issuance," and "method," and clarifies the subject's intent.

[0058] Server: Passes the analysis results to the generative AI model.

[0059] 3. Answer Generation

[0060] Generative AI model: Generates appropriate answers to user inquiries based on the analysis results of the natural language processing engine.

[0061] Generative AI model: Creates a response that includes detailed information and escalation as needed. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID."

[0062] 4. Providing generated answers

[0063] Server: Formats the answers received from the generative AI model and sends them to the device.

[0064] Terminal: Displays the generated answer to the user.

[0065] 5. Receiving Feedback

[0066] Terminal: The user enters feedback on the provided answer and presses the submit button.

[0067] Server: Receives feedback sent from devices and stores it in a database.

[0068] 6. Use of Feedback

[0069] Server: Passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy.

[0070] Specific examples

[0071] For example, when a user types "Please tell me how to issue a resident registration card" into a device and sends it, the server receives this and requests analysis from the natural language processing engine. Keywords such as "resident registration card," "issuance," and "method" are extracted as a result of the analysis. Based on this, the generative AI model generates a response such as "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and sends this to the user's device via the server.

[0072] In this way, the automatic response system based on the present invention can improve the efficiency of inquiries handled by government offices and the quality of services to citizens.

[0073] The processing flow will be explained below.

[0074] Step 1:

[0075] Terminal: The user enters an inquiry on the government website or application and presses the send button. For example, they might enter, "Please tell me how to issue a resident registration card."

[0076] Step 2:

[0077] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[0078] Step 3:

[0079] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[0080] Step 4:

[0081] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[0082] Step 5:

[0083] Server: Sends the analysis results received from the natural language processing engine to the generative AI model. Passes the analysis results to the generative AI model's API as input data.

[0084] Step 6:

[0085] Generative AI model: Generates an appropriate response based on the analysis results. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[0086] Step 7:

[0087] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[0088] Step 8:

[0089] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[0090] Step 9:

[0091] Terminal: The user enters feedback on the provided answer and presses the submit button. Feedback such as "The answer was easy to understand" or "I would like more specific information" is sent.

[0092] Step 10:

[0093] Server: Receives feedback sent from the device and stores it in a database. Passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy.

[0094] Example 1

[0095] 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."

[0096] Conventional government office inquiry response systems have struggled to provide fast and accurate responses to large volumes of inquiries, resulting in labor shortages and increased workloads. They also lacked a system for efficiently utilizing feedback on the answers provided to improve the accuracy of the system. Furthermore, while there is a need to respond to user inquiries in real time, there are also technical challenges in achieving this.

[0097] 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.

[0098] In this invention, the server includes means for receiving inquiries from users, means for saving the received inquiry content in a database, means for analyzing the inquiry content using a natural language processing engine, means for generating an answer using a generative AI model based on the analysis result, means for providing the generated answer to the user terminal, means for receiving user feedback and saving it in a database, and means for using the received feedback as learning data for the generative AI model to improve the accuracy of the answer, thereby enabling improved efficiency and accuracy in responding to inquiries and real-time user responses.

[0099] "Users" refers to citizens and residents who use the government's inquiry system.

[0100] "Inquiries" refer to questions or requests that users enter and submit through government websites or applications.

[0101] "Terminal" refers to a device used by a user to enter and send a query.

[0102] "Server" refers to a central computer system that receives and processes user queries.

[0103] A "natural language processing engine" refers to software or a program for analyzing the content of user inquiries.

[0104] A "generative AI model" refers to an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[0105] "Answer" refers to the appropriate response generated by the generative AI model based on the user's inquiry.

[0106] "Feedback" refers to the evaluation or opinion that a user enters in response to a provided answer.

[0107] A "database" refers to a storage system for storing data such as inquiries, generated responses, and user feedback.

[0108] "Training data" refers to data used to improve the performance of a generative AI model based on received feedback.

[0109] The present invention relates to an automatic response system that accepts inquiries from users, analyzes them, and generates and provides appropriate answers, thereby improving the efficiency of government office operations and improving public services. The following describes specific embodiments of the present invention.

[0110] Overall system configuration

[0111] The system consists of the following main components:

[0112] 1. User Device

[0113] This is a device with an interface that allows citizens to make inquiries using government websites or applications. This device can be any common computing device such as a personal computer (PC), smartphone, or tablet.

[0114] 2. Server

[0115] A central computer system that receives and processes user inquiries, analyzes the content of the inquiry, generates answers, and receives feedback. This server can be a cloud-based server or an on-premise server.

[0116] 3. Natural Language Processing Engine

[0117] This is software and a program for analyzing user inquiries. This engine analyzes the main keywords and intent of the inquiry and passes it on to subsequent processing. Specific software used is a Python natural language processing library (e.g., NLTK or spaCy).

[0118] 4. Generative AI Models

[0119] This is an artificial intelligence model that generates appropriate answers based on the analysis results of a natural language processing engine. This AI model is built using a machine learning framework (e.g., TensorFlow or PyTorch). The generative AI model continuously learns based on user feedback to improve the accuracy of its answers.

[0120] 5. Database

[0121] A storage system for saving data such as inquiry contents, generated answers, and user feedback, etc. It uses a relational database (e.g., MySQL, PostgreSQL) or a NoSQL database (e.g., MongoDB, DynamoDB).

[0122] Program processing

[0123] The program processing of this system is carried out in the following manner.

[0124] 1. Receiving Inquiries

[0125] Terminal: The user enters a question, such as how to issue a resident registration card, through a website or application and presses the send button. For example, the user enters a question such as "How do I issue a resident registration card?"

[0126] Server: Receives the inquiry sent from the user terminal and temporarily stores it in a database.

[0127] 2. Analysis of inquiry content

[0128] Server: Sends the received query content to the natural language processing engine.

[0129] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident registration card," "issuance," and "method," and returns the analysis results to the server.

[0130] 3. Answer Generation

[0131] Server: Passes the analysis results from the natural language processing engine to the generative AI model.

[0132] Generative AI model: Based on the analysis results, it generates a specific answer such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is identification."

[0133] 4. Providing generated answers

[0134] Server: Sends the answer received from the generative AI model to the user's device.

[0135] Terminal: Displays the generated answer to the user, for example, as text on a web page or in the form of a chatbot within an application.

[0136] 5. Receiving Feedback

[0137] Terminal: The user enters feedback on the provided answer and presses the submit button.

[0138] Server: Receives feedback sent from devices and stores it in a database.

[0139] 6. Use of Feedback

[0140] Server: The received feedback is used as learning data for the generative AI model. Based on this feedback data, the generative AI model re-learns and aims to improve the accuracy of its answers.

[0141] Specific examples

[0142] For example, suppose a user types and sends a query on their device, "Please tell me how to issue a resident registration card." The server receives this query and requests analysis from a natural language processing engine. Keywords such as "resident registration card," "issuance," and "method" are extracted as a result of the analysis. Based on this, the generative AI model generates a response: "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This response is sent to the user's device via the server. After seeing this response, the user enters feedback, which is received by the server and stored in a database. The generative AI model then re-learns based on the feedback, allowing it to provide more accurate responses to future inquiries.

[0143] Prompt Sentence Examples

[0144] "A user has submitted the inquiry '[How do I issue a residence certificate?]'. Please generate an appropriate answer based on this information."

[0145] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0146] Step 1:

[0147] Receiving inquiries

[0148] Terminal: The user enters the details of their inquiry (e.g., "Please tell me how to issue a resident registration card") through the government's website or application and presses the send button.

[0149] Input: Text data of the inquiry.

[0150] Server: Receives inquiries sent from user terminals and temporarily stores them in a database.

[0151] Output: The query is saved in the database.

[0152] Step 2:

[0153] Analysis of inquiry content

[0154] Server: Sends the received query content to the natural language processing engine.

[0155] Input: Text data of the inquiry content stored in the database.

[0156] Natural language processing engine: Analyzes the content of the inquiry and extracts keywords such as "resident certificate," "issuance," and "method." The analysis results are returned to the server.

[0157] Output: Analysis results including extracted keywords and query intent.

[0158] Step 3:

[0159] Generate answers

[0160] Server: Passes the analysis results from the natural language processing engine to the generative AI model.

[0161] Input: Analysis result data.

[0162] Generative AI model: Based on the analysis results, it generates appropriate answers to user inquiries. It generates specific answers such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID."

[0163] Output: The text data of the generated answers.

[0164] Step 4:

[0165] Providing generated answers

[0166] Server: Formats the answers received from the generative AI model and sends them to the user's device. To make the answers easier to view, the server formats the data in HTML or JSON format.

[0167] Input: Text data of the answer from the generative AI model.

[0168] Terminal: The generated answer is displayed to the user. For example, the answer can be displayed on a web page or in the form of a chatbot within an application.

[0169] Output: The answer displayed to the user.

[0170] Step 5:

[0171] Receiving Feedback

[0172] Terminal: The user enters feedback (e.g., "helpful" or "not helpful") for the provided answer and presses the submit button. It is also possible to enter specific comments.

[0173] Input: User feedback data.

[0174] Server: Receives feedback sent from devices and stores it in a database.

[0175] Output: Feedback data stored in a database.

[0176] Step 6:

[0177] Using Feedback

[0178] Server: Passes received feedback to the generative AI model.

[0179] Input: Feedback data stored in the database.

[0180] Generative AI model: The model is retrained based on the received feedback data to improve the accuracy of answers and the quality of responses.

[0181] Output: The retrained generative AI model.

[0182] The above is the specific program processing flow of the system, which improves the efficiency and accuracy of inquiries and enables real-time user support.

[0183] (Application example 1)

[0184] 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."

[0185] In modern electronic payment services, it is important to respond quickly and accurately to user inquiries. However, conventional systems take a long time to analyze the inquiry content and generate answers, and the answers provided are often not accurate enough. This makes it difficult to improve user satisfaction. Another problem is that feedback is not fully utilized, hindering system improvement. To solve these issues, there is a need for the development of a system that provides quick and accurate answers and efficiently utilizes feedback as training data.

[0186] 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.

[0187] In this invention, the server includes means for receiving inquiries from users, means for analyzing the received inquiries, means for generating appropriate answers based on the inquiries, means for providing the generated answers to the users, means for receiving feedback on the provided answers, means for using the received feedback as training data, means for identifying inquiries related to payments, and means for generating payment-related information. This enables quick and accurate analysis of inquiries and generation of answers in electronic payment services, thereby improving user satisfaction. Furthermore, by efficiently utilizing feedback as training data, the system can be continuously improved.

[0188] A "user inquiry" is the act of using an electronic device to ask a question or request information or a service.

[0189] "Means for receiving" refers to the functions and methods for obtaining information via electronic devices or communication systems.

[0190] "Means for analyzing the content of inquiries" refers to techniques and methods for understanding the meaning and intent of the information obtained and extracting relevant information.

[0191] "Means for generating appropriate answers" refers to techniques and methods for providing accurate and useful information to users' questions based on the analysis results.

[0192] "Means for providing the generated answer to the user" refers to the method or technology for displaying the generated answer on the user's terminal.

[0193] "Means for receiving feedback" refers to techniques and methods for receiving evaluations and opinions from users.

[0194] "Means of using feedback as training data" refers to techniques and methods for improving system performance based on received evaluations and opinions.

[0195] "Payment Inquiry" means a question or request from a user regarding the payment procedure or process.

[0196] "Means for generating payment information" refers to the technology or method for organizing and providing information related to payment procedures.

[0197] The present invention relates to an automatic response system that accepts and analyzes inquiries from users, and generates and provides appropriate responses, thereby improving the efficiency of electronic payment services and increasing user satisfaction. The following describes specific embodiments of the present invention.

[0198] Overall system configuration

[0199] The system consists of the following main components:

[0200] 1. User Device

[0201] A device with an interface that allows users to make inquiries using an electronic payment service application. Typical examples are smartphones and tablets.

[0202] 2. Server

[0203] A central computer system that receives inquiries, analyzes them, generates answers, and receives feedback. Often runs on a cloud-based server.

[0204] 3. Natural Language Processing Engine

[0205] This is software that analyzes the content of user inquiries and extracts key keywords and intent. Here, we use SpaCy (ja_core_news_sm model).

[0206] 4. Generative AI Models

[0207] This is an AI model that generates appropriate answers based on the analysis results of a natural language processing engine. It uses Huggingface's Transformers library.

[0208] 5. Database

[0209] It is a storage system for saving data such as inquiries, generated answers, and feedback from users.

[0210] System Program Processing

[0211] The server first receives the inquiry sent from the user's device. Next, it sends that content to a natural language processing engine (SpaCy) for analysis. The keywords and intent extracted as a result of the analysis are passed to a generative AI model (Transformers), which generates an appropriate answer. The generated answer is sent back to the user's device and provided to the user. When the user sends feedback on the answer provided, the server receives that feedback and stores it in a database. The received feedback is used as learning data for the generative AI model.

[0212] Specific examples

[0213] For example, if a user types "What should I do if my credit card payment doesn't work?" into a smartphone app and submits it, this inquiry is sent to the server. On the server, SpaCy's natural language processing engine extracts keywords such as "credit card," "payment," "doesn't work," and "how to fix it." Based on the analysis results, the generative AI model generates an answer such as "If your credit card payment doesn't work, please double-check your card information and try again. If the problem persists, please contact customer support," and provides it to the user.

[0214] Prompt Sentence Examples

[0215] What should I do if my credit card payment doesn't work?

[0216] The above is a specific embodiment for carrying out the present invention. This system enables quick and accurate analysis of user inquiries in electronic payment services and provides appropriate answers. Furthermore, by efficiently utilizing feedback as learning data, the system can be continuously improved.

[0217] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0218] Step 1:

[0219] User enters a query

[0220] A user uses a smartphone app for an electronic payment service to enter an inquiry and press the send button. As input, the user enters the question, "What should I do if my credit card payment doesn't go through?". As output, this inquiry is sent to the server.

[0221] Step 2:

[0222] The server receives the query

[0223] The server receives the inquiry sent from the user terminal. The input is text data (question) from the user. The output is to temporarily store the received inquiry in a database.

[0224] Step 3:

[0225] The server sends the query to the natural language processing engine

[0226] The server sends the received query to a natural language processing engine (SpaCy) for analysis. The query stored on the server is used as input. The analysis results (extracted keywords and intent) are obtained as output.

[0227] Step 4:

[0228] The natural language processing engine performs the analysis.

[0229] A natural language processing engine analyzes the content of an inquiry and extracts key keywords and intent. The input is text data (the question). The output is keywords such as "credit card," "payment," "not working," and "solution."

[0230] Step 5:

[0231] The server sends the analysis results to the generated AI model

[0232] The server sends the analysis results of the natural language processing engine to the generative AI model (Transformers). The input is the analysis results (extracted keywords and intent). The output is the generated answer.

[0233] Step 6:

[0234] Generative AI models generate answers

[0235] The generative AI model generates an appropriate answer based on the analysis results. The input is the analysis result, such as "My credit card payment is not working" and "What should I do?" The output is the answer, "If your credit card payment is not working, please double-check your card information and try again. If the problem persists, please contact customer support."

[0236] Step 7:

[0237] The server provides the generated answer to the user

[0238] The server formats the answer received from the generative AI model and sends it to the user's device. The input is the generated answer. The output is the answer displayed on the user's device.

[0239] Step 8:

[0240] Users can provide feedback on the answers provided

[0241] The user inputs feedback for the provided answer and presses the submit button. The input is the feedback content. The output is the feedback sent to the server.

[0242] Step 9:

[0243] Server receives feedback

[0244] The server receives the feedback sent from the user terminal and stores it in a database. As input, there is feedback data from the user. As output, the received feedback is stored in the database.

[0245] Step 10:

[0246] The server sends the feedback as training data to the generative AI model

[0247] The server passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy. The input is the received feedback data. The output is an attempt to improve the performance of the generative AI model.

[0248] 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.

[0249] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates answers to the inquiries, and recognizes the emotions of the users to provide appropriate responses. Hereinafter, embodiments of the present invention will be described in detail.

[0250] Overall system configuration and functions

[0251] The system consists of the following main components:

[0252] 1. User Device

[0253] It is a device with an interface that allows users to make inquiries using government websites and applications.

[0254] 2. Server

[0255] It is a central computer system that receives and processes user queries, analyzes the query, generates answers, provides feedback, and recognizes emotions.

[0256] 3. Natural Language Processing Engine

[0257] This is software or a program that analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to the next process.

[0258] 4. Generative AI Models

[0259] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[0260] 5. Emotion Engine

[0261] It is software or a program that recognizes a user's emotions based on the content of the query or user input, and uses the recognized emotions to generate a response or adjust the tone.

[0262] 6. Database

[0263] It is a storage system for storing data on inquiries, generated answers, user feedback, and recognized emotions.

[0264] Program processing

[0265] The program processing of this system will be explained in natural language below.

[0266] 1. Receiving Inquiries

[0267] Terminal: The user enters an inquiry about how to issue a resident registration card, etc., through the government website or application and presses the send button.

[0268] 2. Analysis of inquiry content

[0269] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[0270] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[0271] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[0272] 3. User Emotion Recognition

[0273] Server: Sends the analysis results received from the natural language processing engine to the emotion engine. Calls the emotion engine's API and sends a request to recognize the user's emotion.

[0274] Emotion engine: Based on the user's input, it recognizes emotions such as excitement, anger, confusion, and joy and returns them to the server.

[0275] 4. Answer Generation

[0276] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model.

[0277] Generative AI model: Based on the emotion recognition results, it generates a response with a tone and content that matches the user's emotions. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[0278] 5. Providing generated answers

[0279] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[0280] 6. Displaying the Answer to the User

[0281] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[0282] 7. Receiving Feedback

[0283] Terminal: The user enters feedback on the provided answer and presses the submit button. Feedback such as "The answer was easy to understand" or "I would like more specific information" is sent.

[0284] 8. Use of Feedback and Emotional Data

[0285] Server: Receives feedback sent from the device and stores it in a database. The received feedback and emotional data are passed to the generative AI model, which uses it to improve future response and response accuracy.

[0286] Specific examples

[0287] For example, a user types, "Please tell me how to issue a resident registration card," and this is sent to the server. The server analyzes the content using a natural language processing engine and extracts keywords. At the same time, the emotion engine recognizes the user's emotion as "confusion." Based on the analysis results and emotion recognition results, the generative AI model generates a response in a tone that alleviates the "confusion," such as, "We will answer your questions about how to issue a resident registration card. Please come to the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This allows the user to resolve the problem with peace of mind.

[0288] In this way, the present invention can recognize the user's emotions and generate appropriate responses, thereby making it possible to improve the efficiency of government office inquiries and further enhance the quality of public services.

[0289] The processing flow will be explained below.

[0290] Step 1:

[0291] Terminal: The user uses the government's website or application to enter an inquiry such as "Please tell me how to issue a resident registration card" and presses the send button.

[0292] Step 2:

[0293] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[0294] Step 3:

[0295] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[0296] Step 4:

[0297] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[0298] Step 5:

[0299] Server: Sends the analysis results received from the natural language processing engine to the emotion engine. Calls the emotion engine's API and sends a request to recognize the user's emotion.

[0300] Step 6:

[0301] Emotion engine: Based on the content of the inquiry, it recognizes the user's emotions such as "confusion," "anger," and "joy," and returns the emotional data to the server.

[0302] Step 7:

[0303] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model. Calls the generative AI model's API and sends a request to generate an appropriate answer.

[0304] Step 8:

[0305] Generative AI model: Based on the analysis results and emotion recognition results, it generates an appropriate answer to the question, "Please tell me how to issue a resident registration card." For example, it generates an answer such as, "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[0306] Step 9:

[0307] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[0308] Step 10:

[0309] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[0310] Step 11:

[0311] Device: The user enters their evaluation and opinion on the provided answer and presses the send button, sending a feedback message such as "The answer was easy to understand" or "I'd like to know more."

[0312] Step 12:

[0313] Server: Receives feedback sent from the device and stores it in a database. Passes the received feedback to the generative AI model and emotion engine, and uses it to improve future inquiry responses and response accuracy.

[0314] Example 2

[0315] 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."

[0316] Conventional automated response systems had the ability to provide appropriate answers to user inquiries, but they were unable to respond in a way that took the user's emotions into consideration. This made improving user satisfaction a challenge. Since the system lacked appropriate responses, particularly for users who were confused or angry, there was a need to further improve the quality of response to inquiries.

[0317] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0318] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the content of the received inquiry, means for using a natural language processing engine to analyze the content of the inquiry, means for recognizing the user's emotions, means for generating an appropriate answer based on the analysis result and the emotion recognition result, means for providing the generated answer to the user, means for receiving feedback on the provided answer, and means for using the received feedback and emotion data as learning data. This enables appropriate responses and answers that take the user's emotions into consideration, improving the quality of inquiry responses and increasing user satisfaction.

[0319] "Means for receiving inquiries from users" refers to a function that allows the server to receive the content of inquiries entered by users through the government's website or application.

[0320] The "means for analyzing the content of the received inquiry" is a function that analyzes the content of the user's inquiry received by the server in order to understand it.

[0321] "Means for using a natural language processing engine" refers to a function that uses a software engine that performs natural language processing when analyzing the content of a received inquiry.

[0322] The "means for recognizing the user's emotions" is a function for identifying the emotional state at the time based on the content of the inquiry sent by the user.

[0323] "Means for generating appropriate answers based on analysis results and emotion recognition results" refers to a function for generating appropriate answers based on the results of natural language processing and emotion recognition.

[0324] "Means for providing the generated answer to the user" refers to the function by which the server sends the answer created by the generative AI model to the user's terminal and displays it.

[0325] The "means for receiving feedback on the provided answer" is a function that allows a user to input opinions and evaluations on the provided answer, and the server receives them.

[0326] The "means for using received feedback and emotion data as learning data" is a function for using the feedback and emotion recognition results received from the user to improve the system's performance and response accuracy.

[0327] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates answers to the inquiries, and recognizes the emotions of the users to provide appropriate responses. Hereinafter, embodiments of the present invention will be described in detail.

[0328] Overall system configuration and functions

[0329] The system consists of the following main components:

[0330] 1. User terminal: A device with an interface that allows users to make inquiries using government websites and applications. This includes PCs, smartphones, tablets, etc.

[0331] 2. Server: A central computer system that receives user queries, analyzes them, and generates answers. The server uses the following software and programs:

[0332] Natural language processing engines (e.g., SpaCy and BERT)

[0333] Emotion engine (e.g., IBM Watson Tone Analyzer)

[0334] Generative AI models (e.g., GPT-3)

[0335] Database (e.g. MySQL)

[0336] 3. Natural language processing engine: Software or a program that analyzes user inquiries, extracts the intent and keywords of the inquiry, and passes them on to the next process.

[0337] 4. Generative AI model: An artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of the natural language processing engine.

[0338] 5. Emotion Engine: Software or a program that recognizes the user's emotions based on the query or user input. The recognized emotions are used to generate responses and adjust the tone.

[0339] 6. Database: A storage system for storing data on queries, generated answers, user feedback, and recognized emotions.

[0340] Program processing flow

[0341] 1. Receiving Inquiries

[0342] Device: The user opens the government website or application, enters "Please tell me how to issue a resident registration card" into the inquiry form, and presses the send button.

[0343] 2. Analysis of inquiry content

[0344] Server: Receives the inquiry sent by the user and temporarily stores it in a local database. The stored inquiry is assigned a unique identification number (e.g., UUID).

[0345] Server: Sends the received query content to a natural language processing engine (e.g., SpaCy or BERT) by sending an HTTP POST request to the API endpoint of the natural language processing engine.

[0346] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[0347] 3. User Emotion Recognition

[0348] Server: Sends the analysis results received from the natural language processing engine to the emotion engine (e.g., IBM Watson Tone Analyzer). Sends an HTTP POST request to the emotion engine's API endpoint.

[0349] Emotion engine: Based on the user's input, it recognizes emotions such as excitement, anger, confusion, and joy and returns them to the server.

[0350] 4. Answer Generation

[0351] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model (e.g., GPT-3).

[0352] Generative AI model: Based on the emotion recognition results, it generates a response with a tone and content that matches the user's emotions. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[0353] 5. Providing generated answers

[0354] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[0355] 6. Displaying the Answer to the User

[0356] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[0357] Specific examples

[0358] For example, a user types, "Please tell me how to issue a resident registration card," and this is sent to the server. The server uses a natural language processing engine to analyze the content and extract keywords. At the same time, the emotion engine recognizes the user's emotion as "confusion." Based on this analysis result and the emotion recognition result, the generative AI model generates a response in a tone that alleviates the "confusion," such as, "We will answer your questions about how to issue a resident registration card. Please come to the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This allows the user to resolve the problem with peace of mind.

[0359] Prompt Sentence Examples

[0360] example:

[0361] User's question: "How do I get a certificate of residence issued?"

[0362] Analysis results: {"keywords": ["resident card", "issuance", "method"], "intent": "information_request"}

[0363] Emotion recognition result: "confused"

[0364] Based on this prompt, a generative AI model generates an appropriate response, and the system as a whole accumulates feedback and sentiment data that can be used to improve future inquiries.

[0365] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0366] Step 1:

[0367] The user opens the city hall's website or application, enters "Please tell me how to issue a resident registration card" into the inquiry form, and presses the send button.

[0368] Input: User's inquiry: "Please tell me how to issue a resident registration card."

[0369] Specific operation: When the user clicks the "Send" button, the terminal sends the input contents to the server as an HTTP POST request.

[0370] Output: The query received by the server.

[0371] Step 2:

[0372] The server receives the query sent by the user and temporarily stores it in a local database, where it is assigned a unique identification number (e.g., UUID).

[0373] Input: User's inquiry: "Please tell me how to issue a resident registration card."

[0374] Specific operation: The server saves the inquiry in the "inquiries" table of the MySQL database and issues a unique ID.

[0375] Output: The query and its unique identification number stored in the database.

[0376] Step 3:

[0377] The server sends the received query to a natural language processing engine.

[0378] Input: The query and its unique identification number stored in the database.

[0379] Specific operation: The server converts the query content into JSON format and sends an HTTP POST request to the API endpoint of a natural language processing engine (e.g., SpaCy or BERT).

[0380] Output: The results of the analysis by the natural language processing engine are received in JSON format.

[0381] Step 4:

[0382] The natural language processing engine analyzes the inquiry, extracts keywords such as "resident registration," "issuance," and "method," and analyzes the sentence structure.

[0383] Input: JSON data of the query sent from the server.

[0384] Specific operation: The natural language processing engine performs text analysis, compiles the extracted keywords and intent into JSON format, and returns it to the server.

[0385] Output: JSON format data containing the parsed results.

[0386] Step 5:

[0387] The server sends the analysis results from the natural language processing engine to the emotion engine.

[0388] Input: JSON data of the analysis results obtained from the natural language processing engine.

[0389] Specific operation: The server sends the analysis results to the API endpoint of the emotion engine (e.g., IBM Watson Tone Analyzer) via an HTTP POST request.

[0390] Output: Receives emotion data recognized by the emotion engine in JSON format.

[0391] Step 6:

[0392] The emotion engine recognizes the user's emotions based on the content of the query and identifies emotions such as "confusion."

[0393] Input: JSON data of the analysis result sent from the server.

[0394] Specific operation: The emotion engine performs emotion analysis and returns the recognized emotion tag to the server.

[0395] Output: JSON formatted data containing emotion recognition results.

[0396] Step 7:

[0397] The server sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model.

[0398] Input: JSON data of analysis results and emotion recognition results obtained from the natural language processing engine and emotion engine.

[0399] Specific operation: The server sends a prompt message to the API endpoint of the generative AI model (e.g., GPT-3) based on the analysis results and emotion recognition results.

[0400] Output: JSON-formatted data containing the answers generated by the generative AI model.

[0401] Step 8:

[0402] The generative AI model generates responses with a tone and content that matches the user's emotions based on the emotion recognition results.

[0403] Input: Prompt sentence based on analysis results and emotion recognition results.

[0404] Specific operation: The generative AI model generates an appropriate answer and returns it to the server in JSON format.

[0405] Output: JSON-formatted data containing the appropriate answer from the generative AI model.

[0406] Step 9:

[0407] The server formats the answers received from the generative AI model and converts them into data for transmission to the user's device.

[0408] Input: JSON data of the appropriate answer data from the generative AI model.

[0409] What happens: The server embeds the response in an HTML template and reformats it for display in the user's browser.

[0410] Output: Formatted response data.

[0411] Step 10:

[0412] The user terminal receives the response data sent from the server and displays it on the user interface.

[0413] Input: Formatted response data.

[0414] Specific operation: The web browser on the user's device renders the received HTML data and displays it to the user.

[0415] Output: Answers displayed on the screen.

[0416] Step 11:

[0417] The user enters feedback on the provided answer and presses the submit button.

[0418] Input: The feedback entered by the user.

[0419] Specific operation: When a user fills out the feedback form and presses the submit button, the terminal sends the feedback to the server as an HTTP POST request.

[0420] Output: The feedback received by the server.

[0421] Step 12:

[0422] The server receives the feedback sent from the device and stores it in a database. The received feedback and emotional data are used as training data for the generative AI model.

[0423] Input: User feedback.

[0424] What it does: The server stores the feedback in a "feedback" table in the database and uses it for subsequent analysis and improvement processes.

[0425] Output: Feedback content and sentiment data stored in a database.

[0426] (Application example 2)

[0427] 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."

[0428] Conventional query processing systems do not recognize the user's emotions and generate answers based on them, but simply respond based on keywords, which can result in users not receiving satisfactory answers. Furthermore, when a user requests recommendations for content such as movies or TV dramas, there is a problem in that appropriate recommendations based on the user's emotions cannot be made.

[0429] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the received inquiry content, means for recognizing the user's emotions based on the analyzed content, means for generating an appropriate answer according to the user's emotions, means for providing the generated answer to the user, means for receiving feedback on the provided answer, means for using the received feedback as learning data, and means for recommending movies and TV shows. This makes it possible to respond to inquiries and recommend appropriate movies and TV shows that take the user's emotions into consideration.

[0430] "Means for receiving inquiries from users" refers to a function for receiving questions or requests entered by users through devices such as smartphones or computers from those devices.

[0431] The "means for analyzing the content of the received inquiry" is a function for understanding the content of the inquiry received from the user and performing processing to extract appropriate information.

[0432] The "means for recognizing the user's emotions based on the analyzed content" is a function for identifying the emotions the user is feeling based on the analysis results of the inquiry content.

[0433] The "means for generating an appropriate response according to the user's emotions" is a function that takes into account the recognized emotions of the user and creates an answer with a tone and content that is appropriate to those emotions.

[0434] The "means for providing the generated answer to the user" is a function for displaying or replying to the generated answer on the user's device.

[0435] The "means for receiving feedback on the provided answer" is a function that allows a user to input and receive an evaluation or additional comments on the provided answer.

[0436] "Means of using received feedback as learning data" is a function that accumulates feedback from users and uses it to improve the accuracy of future inquiries and to improve the system.

[0437] The "means for recommending movies and TV dramas" is a function for recommending appropriate movies and TV dramas to users based on the user's emotions and inquiries.

[0438] MODE FOR CARRYING OUT THE INVENTION

[0439] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates responses to the inquiries, and further recognizes the emotions of the users to provide appropriate responses. Hereinafter, specific embodiments of the present invention will be described.

[0440] Overall system configuration and functions

[0441] The system consists of the following main components:

[0442] 1. User Device:

[0443] It is a device with an interface that allows users to make inquiries through a smartphone application.

[0444] Hardware: Smartphone (e.g. iPhone, Android device)

[0445] Software: User Interface Application

[0446] 2. Server:

[0447] It is a central computer system that receives and processes user queries, analyzes the query, generates answers, provides feedback, and recognizes emotions.

[0448] Hardware: Server computer

[0449] Software: Web server, database server

[0450] 3. Natural Language Processing Engine:

[0451] This software analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to the next process.

[0452] Usage examples: SpaCy, NLTK

[0453] 4. Generative AI Models:

[0454] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[0455] Usage example: GPT-4, BERT

[0456] 5. Emotion Engine:

[0457] This software recognizes user emotions based on the content of the query and user input, and uses the recognized emotions to generate responses and adjust tone.

[0458] Usage examples: IBM Watson Tone Analyzer, Microsoft Azure Text Analytics

[0459] 6. Database:

[0460] It is a storage system for storing data on inquiries, generated answers, user feedback, and recognized emotions.

[0461] Usage example: MySQL, PostgreSQL

[0462] 7. Movie and TV series recommendation module:

[0463] This software recommends appropriate movies and dramas based on the user's emotions and inquiries.

[0464] Program processing

[0465] This program operates as follows:

[0466] 1. Receiving Inquiries

[0467] The user terminal receives an inquiry from the user through a smartphone app, which is then sent to the server.

[0468] 2. Analysis of inquiry content

[0469] The server sends the received inquiry to a natural language processing engine, which analyzes the inquiry and extracts keywords.

[0470] 3. User Emotion Recognition

[0471] The server receives the analysis results from the natural language processing engine and sends them to the emotion engine, which recognizes the user's emotions and returns the results to the server.

[0472] 4. Answer Generation

[0473] The server sends the analysis results and emotion recognition results to a generative AI model, which generates an answer appropriate to the user's emotions. For example, if the system recognizes that the user is "depressed," it will "recommend movies and TV shows that will help alleviate the depression."

[0474] Prompt Sentence Examples

[0475] If the user is in an emotionally depressed state, recommend movies or TV shows based on the following: "I'm feeling down. Can you recommend some movies that will cheer me up?"

[0476] 5. Providing generated answers

[0477] The server formats the generated answer, converts it into data for display on the user's device, and transmits it to the user. The user can view the answer through the user interface on their smartphone.

[0478] 6. Receive feedback and learn

[0479] The user terminal receives feedback from the user regarding the provided answer and sends it to the server, which stores this feedback in a database and uses it to improve future responses to inquiries.

[0480] In this way, the present invention can recognize the user's emotions and generate appropriate responses, thereby making it possible to improve the efficiency of inquiry response work and user satisfaction.

[0481] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0482] Step 1:

[0483] The user uses a smartphone app to input the inquiry and press the send button. An example input is "I'm feeling down, can you recommend a movie that will cheer me up?" The device then sends this inquiry to the server. The input data is the inquiry in text format, and the output is the data transferred to the server.

[0484] Step 2:

[0485] The server sends the received query content to a natural language processing engine, which analyzes the content and extracts key keywords (e.g., "feeling depressed," "cheerful," "movies"). The input data is the user's query content, and the output data is the analysis results including the extracted keywords.

[0486] Step 3:

[0487] The server sends the analysis results received from the natural language processing engine to the emotion engine. The emotion engine recognizes the user's emotions based on these analysis results. For example, it identifies emotions such as "feeling depressed." The input data is the analysis results, and the output data is the recognized emotional information.

[0488] Step 4:

[0489] The server sends the analysis results and emotion recognition results to the generative AI model. The generative AI model generates an answer that is appropriate to the user's emotions based on these inputs. For example, if the emotion is recognized as "depressed," it generates a recommendation such as "comedy movies" as an "uplifting movie." The input data are the analysis results and emotion recognition results, and the output data is the generated answer. The specific operation uses the following prompt sentence:

[0490] If the user is in an emotionally depressed state, recommend movies or TV shows based on the following: "I'm feeling down. Can you recommend some movies that will cheer me up?"

[0491] Step 5:

[0492] The server formats the generated answer and converts it into an appropriate data format (e.g., JSON, XML, HTML) for transmission to the device. The input data is the generated answer, and the output data is the formatted answer.

[0493] Step 6:

[0494] The terminal receives the formatted answer data sent from the server and displays it on the user interface. The user can check recommendations such as "Title A," which is a comedy movie, on the screen. The input data is the formatted answer data, and the output data is the displayed answer.

[0495] Step 7:

[0496] The user inputs and transmits feedback on the provided answer through the terminal. For example, the feedback may include "The recommendation for this movie was accurate." The terminal transmits this feedback to the server. The input data is the user's feedback, and the output data is the feedback transmitted to the server.

[0497] Step 8:

[0498] The server stores the feedback received from the device in a database and associates it with the analysis results and emotion recognition results. This feedback is used to improve future inquiries. The input data is the user feedback, and the output data is the stored feedback information.

[0499] 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.

[0500] 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.

[0501] 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.

[0502] [Second embodiment]

[0503] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0504] 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.

[0505] 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).

[0506] 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.

[0507] 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.

[0508] 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).

[0509] 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.

[0510] 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.

[0511] 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.

[0512] 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.

[0513] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0514] 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."

[0515] The present invention relates to an automatic response system that accepts inquiries from users, analyzes them, and generates and provides appropriate answers, thereby improving the efficiency of government office operations and improving public services. The following describes specific embodiments of the present invention.

[0516] Overall system configuration

[0517] The system consists of the following main components:

[0518] 1. User Device

[0519] It is a device that has an interface that allows citizens to make inquiries using government websites or applications.

[0520] 2. Server

[0521] A central computer system that receives and processes user inquiries, including analyzing the inquiry, generating answers, and receiving feedback.

[0522] 3. Natural Language Processing Engine

[0523] This is software or a program that analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to subsequent processing.

[0524] 4. Generative AI Models

[0525] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[0526] 5. Database

[0527] It is a storage system for saving data such as inquiry details, generated answers, and feedback from users.

[0528] Program processing

[0529] The program processing of this system will be explained in natural language below.

[0530] 1. Receiving Inquiries

[0531] Terminal: The user enters an inquiry about how to issue a resident registration card, etc., through the government website or application and presses the send button.

[0532] Server: Receives the inquiry sent from the terminal and temporarily stores it in a database.

[0533] 2. Analysis of inquiry content

[0534] Server: Sends the received query content to the natural language processing engine.

[0535] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident registration card," "issuance," and "method," and clarifies the subject's intent.

[0536] Server: Passes the analysis results to the generative AI model.

[0537] 3. Answer Generation

[0538] Generative AI model: Generates appropriate answers to user inquiries based on the analysis results of the natural language processing engine.

[0539] Generative AI model: Creates a response that includes detailed information and escalation as needed. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID."

[0540] 4. Providing generated answers

[0541] Server: Formats the answers received from the generative AI model and sends them to the device.

[0542] Terminal: Displays the generated answer to the user.

[0543] 5. Receiving Feedback

[0544] Terminal: The user enters feedback on the provided answer and presses the submit button.

[0545] Server: Receives feedback sent from devices and stores it in a database.

[0546] 6. Use of Feedback

[0547] Server: Passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy.

[0548] Specific examples

[0549] For example, when a user types "Please tell me how to issue a resident registration card" into a device and sends it, the server receives this and requests analysis from the natural language processing engine. Keywords such as "resident registration card," "issuance," and "method" are extracted as a result of the analysis. Based on this, the generative AI model generates a response such as "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and sends this to the user's device via the server.

[0550] In this way, the automatic response system based on the present invention can improve the efficiency of inquiries handled by government offices and the quality of services to citizens.

[0551] The processing flow will be explained below.

[0552] Step 1:

[0553] Terminal: The user enters an inquiry on the government website or application and presses the send button. For example, they might enter, "Please tell me how to issue a resident registration card."

[0554] Step 2:

[0555] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[0556] Step 3:

[0557] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[0558] Step 4:

[0559] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[0560] Step 5:

[0561] Server: Sends the analysis results received from the natural language processing engine to the generative AI model. Passes the analysis results to the generative AI model's API as input data.

[0562] Step 6:

[0563] Generative AI model: Generates an appropriate response based on the analysis results. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[0564] Step 7:

[0565] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[0566] Step 8:

[0567] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[0568] Step 9:

[0569] Terminal: The user enters feedback on the provided answer and presses the submit button. Feedback such as "The answer was easy to understand" or "I would like more specific information" is sent.

[0570] Step 10:

[0571] Server: Receives feedback sent from the device and stores it in a database. Passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy.

[0572] Example 1

[0573] 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."

[0574] Conventional government office inquiry response systems have struggled to provide fast and accurate responses to large volumes of inquiries, resulting in labor shortages and increased workloads. They also lacked a system for efficiently utilizing feedback on the answers provided to improve the accuracy of the system. Furthermore, while there is a need to respond to user inquiries in real time, there are also technical challenges in achieving this.

[0575] 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.

[0576] In this invention, the server includes means for receiving inquiries from users, means for saving the received inquiry content in a database, means for analyzing the inquiry content using a natural language processing engine, means for generating an answer using a generative AI model based on the analysis result, means for providing the generated answer to the user terminal, means for receiving user feedback and saving it in a database, and means for using the received feedback as learning data for the generative AI model to improve the accuracy of the answer, thereby enabling improved efficiency and accuracy in responding to inquiries and real-time user responses.

[0577] "Users" refers to citizens and residents who use the government's inquiry system.

[0578] "Inquiries" refer to questions or requests that users enter and submit through government websites or applications.

[0579] "Terminal" refers to a device used by a user to enter and send a query.

[0580] "Server" refers to a central computer system that receives and processes user queries.

[0581] A "natural language processing engine" refers to software or a program for analyzing the content of user inquiries.

[0582] A "generative AI model" refers to an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[0583] "Answer" refers to the appropriate response generated by the generative AI model based on the user's inquiry.

[0584] "Feedback" refers to the evaluation or opinion that a user enters in response to a provided answer.

[0585] A "database" refers to a storage system for storing data such as inquiries, generated responses, and user feedback.

[0586] "Training data" refers to data used to improve the performance of a generative AI model based on received feedback.

[0587] The present invention relates to an automatic response system that accepts inquiries from users, analyzes them, and generates and provides appropriate answers, thereby improving the efficiency of government office operations and improving public services. The following describes specific embodiments of the present invention.

[0588] Overall system configuration

[0589] The system consists of the following main components:

[0590] 1. User Device

[0591] This is a device with an interface that allows citizens to make inquiries using government websites or applications. This device can be any common computing device such as a personal computer (PC), smartphone, or tablet.

[0592] 2. Server

[0593] A central computer system that receives and processes user inquiries, analyzes the content of the inquiry, generates answers, and receives feedback. This server can be a cloud-based server or an on-premise server.

[0594] 3. Natural Language Processing Engine

[0595] This is software and a program for analyzing user inquiries. This engine analyzes the main keywords and intent of the inquiry and passes it on to subsequent processing. Specific software used is a Python natural language processing library (e.g., NLTK or spaCy).

[0596] 4. Generative AI Models

[0597] This is an artificial intelligence model that generates appropriate answers based on the analysis results of a natural language processing engine. This AI model is built using a machine learning framework (e.g., TensorFlow or PyTorch). The generative AI model continuously learns based on user feedback to improve the accuracy of its answers.

[0598] 5. Database

[0599] A storage system for saving data such as inquiry contents, generated answers, and user feedback, etc. It uses a relational database (e.g., MySQL, PostgreSQL) or a NoSQL database (e.g., MongoDB, DynamoDB).

[0600] Program processing

[0601] The program processing of this system is carried out in the following manner.

[0602] 1. Receiving Inquiries

[0603] Terminal: The user enters a question, such as how to issue a resident registration card, through a website or application and presses the send button. For example, the user enters a question such as "How do I issue a resident registration card?"

[0604] Server: Receives the inquiry sent from the user terminal and temporarily stores it in a database.

[0605] 2. Analysis of inquiry content

[0606] Server: Sends the received query content to the natural language processing engine.

[0607] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident registration card," "issuance," and "method," and returns the analysis results to the server.

[0608] 3. Answer Generation

[0609] Server: Passes the analysis results from the natural language processing engine to the generative AI model.

[0610] Generative AI model: Based on the analysis results, it generates a specific answer such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is identification."

[0611] 4. Providing generated answers

[0612] Server: Sends the answer received from the generative AI model to the user's device.

[0613] Terminal: Displays the generated answer to the user, for example, as text on a web page or in the form of a chatbot within an application.

[0614] 5. Receiving Feedback

[0615] Terminal: The user enters feedback on the provided answer and presses the submit button.

[0616] Server: Receives feedback sent from devices and stores it in a database.

[0617] 6. Use of Feedback

[0618] Server: The received feedback is used as learning data for the generative AI model. Based on this feedback data, the generative AI model re-learns and aims to improve the accuracy of its answers.

[0619] Specific examples

[0620] For example, suppose a user types and sends a query on their device, "Please tell me how to issue a resident registration card." The server receives this query and requests analysis from a natural language processing engine. Keywords such as "resident registration card," "issuance," and "method" are extracted as a result of the analysis. Based on this, the generative AI model generates a response: "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This response is sent to the user's device via the server. After seeing this response, the user enters feedback, which is received by the server and stored in a database. The generative AI model then re-learns based on the feedback, allowing it to provide more accurate responses to future inquiries.

[0621] Prompt Sentence Examples

[0622] "A user has submitted the inquiry '[How do I issue a residence certificate?]'. Please generate an appropriate answer based on this information."

[0623] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0624] Step 1:

[0625] Receiving inquiries

[0626] Terminal: The user enters the details of their inquiry (e.g., "Please tell me how to issue a resident registration card") through the government's website or application and presses the send button.

[0627] Input: Text data of the inquiry.

[0628] Server: Receives inquiries sent from user terminals and temporarily stores them in a database.

[0629] Output: The query is saved in the database.

[0630] Step 2:

[0631] Analysis of inquiry content

[0632] Server: Sends the received query content to the natural language processing engine.

[0633] Input: Text data of the inquiry content stored in the database.

[0634] Natural language processing engine: Analyzes the content of the inquiry and extracts keywords such as "resident certificate," "issuance," and "method." The analysis results are returned to the server.

[0635] Output: Analysis results including extracted keywords and query intent.

[0636] Step 3:

[0637] Generate answers

[0638] Server: Passes the analysis results from the natural language processing engine to the generative AI model.

[0639] Input: Analysis result data.

[0640] Generative AI model: Based on the analysis results, it generates appropriate answers to user inquiries. It generates specific answers such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID."

[0641] Output: The text data of the generated answers.

[0642] Step 4:

[0643] Providing generated answers

[0644] Server: Formats the answers received from the generative AI model and sends them to the user's device. To make the answers easier to view, the server formats the data in HTML or JSON format.

[0645] Input: Text data of the answer from the generative AI model.

[0646] Terminal: The generated answer is displayed to the user. For example, the answer can be displayed on a web page or in the form of a chatbot within an application.

[0647] Output: The answer displayed to the user.

[0648] Step 5:

[0649] Receiving Feedback

[0650] Terminal: The user enters feedback (e.g., "helpful" or "not helpful") for the provided answer and presses the submit button. It is also possible to enter specific comments.

[0651] Input: User feedback data.

[0652] Server: Receives feedback sent from devices and stores it in a database.

[0653] Output: Feedback data stored in a database.

[0654] Step 6:

[0655] Using Feedback

[0656] Server: Passes received feedback to the generative AI model.

[0657] Input: Feedback data stored in the database.

[0658] Generative AI model: The model is retrained based on the received feedback data to improve the accuracy of answers and the quality of responses.

[0659] Output: The retrained generative AI model.

[0660] The above is the specific program processing flow of the system, which improves the efficiency and accuracy of inquiries and enables real-time user support.

[0661] (Application example 1)

[0662] 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."

[0663] In modern electronic payment services, it is important to respond quickly and accurately to user inquiries. However, conventional systems take a long time to analyze the inquiry content and generate answers, and the answers provided are often not accurate enough. This makes it difficult to improve user satisfaction. Another problem is that feedback is not fully utilized, hindering system improvement. To solve these issues, there is a need for the development of a system that provides quick and accurate answers and efficiently utilizes feedback as training data.

[0664] 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.

[0665] In this invention, the server includes means for receiving inquiries from users, means for analyzing the received inquiries, means for generating appropriate answers based on the inquiries, means for providing the generated answers to the users, means for receiving feedback on the provided answers, means for using the received feedback as training data, means for identifying inquiries related to payments, and means for generating payment-related information. This enables quick and accurate analysis of inquiries and generation of answers in electronic payment services, thereby improving user satisfaction. Furthermore, by efficiently utilizing feedback as training data, the system can be continuously improved.

[0666] A "user inquiry" is the act of using an electronic device to ask a question or request information or a service.

[0667] "Means for receiving" refers to the functions and methods for obtaining information via electronic devices or communication systems.

[0668] "Means for analyzing the content of inquiries" refers to techniques and methods for understanding the meaning and intent of the information obtained and extracting relevant information.

[0669] "Means for generating appropriate answers" refers to techniques and methods for providing accurate and useful information to users' questions based on the analysis results.

[0670] "Means for providing the generated answer to the user" refers to the method or technology for displaying the generated answer on the user's terminal.

[0671] "Means for receiving feedback" refers to techniques and methods for receiving evaluations and opinions from users.

[0672] "Means of using feedback as training data" refers to techniques and methods for improving system performance based on received evaluations and opinions.

[0673] "Payment Inquiry" means a question or request from a user regarding the payment procedure or process.

[0674] "Means for generating payment information" refers to the technology or method for organizing and providing information related to payment procedures.

[0675] The present invention relates to an automatic response system that accepts and analyzes inquiries from users, and generates and provides appropriate responses, thereby improving the efficiency of electronic payment services and increasing user satisfaction. The following describes specific embodiments of the present invention.

[0676] Overall system configuration

[0677] The system consists of the following main components:

[0678] 1. User Device

[0679] A device with an interface that allows users to make inquiries using an electronic payment service application. Typical examples are smartphones and tablets.

[0680] 2. Server

[0681] A central computer system that receives inquiries, analyzes them, generates answers, and receives feedback. Often runs on a cloud-based server.

[0682] 3. Natural Language Processing Engine

[0683] This is software that analyzes the content of user inquiries and extracts key keywords and intent. Here, we use SpaCy (ja_core_news_sm model).

[0684] 4. Generative AI Models

[0685] This is an AI model that generates appropriate answers based on the analysis results of a natural language processing engine. It uses Huggingface's Transformers library.

[0686] 5. Database

[0687] It is a storage system for saving data such as inquiries, generated answers, and feedback from users.

[0688] System Program Processing

[0689] The server first receives the inquiry sent from the user's device. Next, it sends that content to a natural language processing engine (SpaCy) for analysis. The keywords and intent extracted as a result of the analysis are passed to a generative AI model (Transformers), which generates an appropriate answer. The generated answer is sent back to the user's device and provided to the user. When the user sends feedback on the answer provided, the server receives that feedback and stores it in a database. The received feedback is used as learning data for the generative AI model.

[0690] Specific examples

[0691] For example, if a user types "What should I do if my credit card payment doesn't work?" into a smartphone app and submits it, this inquiry is sent to the server. On the server, SpaCy's natural language processing engine extracts keywords such as "credit card," "payment," "doesn't work," and "how to fix it." Based on the analysis results, the generative AI model generates an answer such as "If your credit card payment doesn't work, please double-check your card information and try again. If the problem persists, please contact customer support," and provides it to the user.

[0692] Prompt Sentence Examples

[0693] What should I do if my credit card payment doesn't work?

[0694] The above is a specific embodiment for carrying out the present invention. This system enables quick and accurate analysis of user inquiries in electronic payment services and provides appropriate answers. Furthermore, by efficiently utilizing feedback as learning data, the system can be continuously improved.

[0695] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0696] Step 1:

[0697] User enters a query

[0698] A user uses a smartphone app for an electronic payment service to enter an inquiry and press the send button. As input, the user enters the question, "What should I do if my credit card payment doesn't go through?". As output, this inquiry is sent to the server.

[0699] Step 2:

[0700] The server receives the query

[0701] The server receives the inquiry sent from the user terminal. The input is text data (question) from the user. The output is to temporarily store the received inquiry in a database.

[0702] Step 3:

[0703] The server sends the query to the natural language processing engine

[0704] The server sends the received query to a natural language processing engine (SpaCy) for analysis. The query stored on the server is used as input. The analysis results (extracted keywords and intent) are obtained as output.

[0705] Step 4:

[0706] The natural language processing engine performs the analysis.

[0707] A natural language processing engine analyzes the content of an inquiry and extracts key keywords and intent. The input is text data (the question). The output is keywords such as "credit card," "payment," "not working," and "solution."

[0708] Step 5:

[0709] The server sends the analysis results to the generated AI model

[0710] The server sends the analysis results of the natural language processing engine to the generative AI model (Transformers). The input is the analysis results (extracted keywords and intent). The output is the generated answer.

[0711] Step 6:

[0712] Generative AI models generate answers

[0713] The generative AI model generates an appropriate answer based on the analysis results. The input is the analysis result, such as "My credit card payment is not working" and "What should I do?" The output is the answer, "If your credit card payment is not working, please double-check your card information and try again. If the problem persists, please contact customer support."

[0714] Step 7:

[0715] The server provides the generated answer to the user

[0716] The server formats the answer received from the generative AI model and sends it to the user's device. The input is the generated answer. The output is the answer displayed on the user's device.

[0717] Step 8:

[0718] Users can provide feedback on the answers provided

[0719] The user inputs feedback for the provided answer and presses the submit button. The input is the feedback content. The output is the feedback sent to the server.

[0720] Step 9:

[0721] Server receives feedback

[0722] The server receives the feedback sent from the user terminal and stores it in a database. As input, there is feedback data from the user. As output, the received feedback is stored in the database.

[0723] Step 10:

[0724] The server sends the feedback as training data to the generative AI model

[0725] The server passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy. The input is the received feedback data. The output is an attempt to improve the performance of the generative AI model.

[0726] 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.

[0727] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates answers to the inquiries, and recognizes the emotions of the users to provide appropriate responses. Hereinafter, embodiments of the present invention will be described in detail.

[0728] Overall system configuration and functions

[0729] The system consists of the following main components:

[0730] 1. User Device

[0731] It is a device with an interface that allows users to make inquiries using government websites and applications.

[0732] 2. Server

[0733] It is a central computer system that receives and processes user queries, analyzes the query, generates answers, provides feedback, and recognizes emotions.

[0734] 3. Natural Language Processing Engine

[0735] This is software or a program that analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to the next process.

[0736] 4. Generative AI Models

[0737] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[0738] 5. Emotion Engine

[0739] It is software or a program that recognizes a user's emotions based on the content of the query or user input, and uses the recognized emotions to generate a response or adjust the tone.

[0740] 6. Database

[0741] It is a storage system for storing data on inquiries, generated answers, user feedback, and recognized emotions.

[0742] Program processing

[0743] The program processing of this system will be explained in natural language below.

[0744] 1. Receiving Inquiries

[0745] Terminal: The user enters an inquiry about how to issue a resident registration card, etc., through the government website or application and presses the send button.

[0746] 2. Analysis of inquiry content

[0747] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[0748] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[0749] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[0750] 3. User Emotion Recognition

[0751] Server: Sends the analysis results received from the natural language processing engine to the emotion engine. Calls the emotion engine's API and sends a request to recognize the user's emotion.

[0752] Emotion engine: Based on the user's input, it recognizes emotions such as excitement, anger, confusion, and joy and returns them to the server.

[0753] 4. Answer Generation

[0754] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model.

[0755] Generative AI model: Based on the emotion recognition results, it generates a response with a tone and content that matches the user's emotions. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[0756] 5. Providing generated answers

[0757] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[0758] 6. Displaying the Answer to the User

[0759] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[0760] 7. Receiving Feedback

[0761] Terminal: The user enters feedback on the provided answer and presses the submit button. Feedback such as "The answer was easy to understand" or "I would like more specific information" is sent.

[0762] 8. Use of Feedback and Emotional Data

[0763] Server: Receives feedback sent from the device and stores it in a database. The received feedback and emotional data are passed to the generative AI model, which uses it to improve future response and response accuracy.

[0764] Specific examples

[0765] For example, a user types, "Please tell me how to issue a resident registration card," and this is sent to the server. The server analyzes the content using a natural language processing engine and extracts keywords. At the same time, the emotion engine recognizes the user's emotion as "confusion." Based on the analysis results and emotion recognition results, the generative AI model generates a response in a tone that alleviates the "confusion," such as, "We will answer your questions about how to issue a resident registration card. Please come to the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This allows the user to resolve the problem with peace of mind.

[0766] In this way, the present invention can recognize the user's emotions and generate appropriate responses, thereby making it possible to improve the efficiency of government office inquiries and further enhance the quality of public services.

[0767] The processing flow will be explained below.

[0768] Step 1:

[0769] Terminal: The user uses the government's website or application to enter an inquiry such as "Please tell me how to issue a resident registration card" and presses the send button.

[0770] Step 2:

[0771] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[0772] Step 3:

[0773] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[0774] Step 4:

[0775] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[0776] Step 5:

[0777] Server: Sends the analysis results received from the natural language processing engine to the emotion engine. Calls the emotion engine's API and sends a request to recognize the user's emotion.

[0778] Step 6:

[0779] Emotion engine: Based on the content of the inquiry, it recognizes the user's emotions such as "confusion," "anger," and "joy," and returns the emotional data to the server.

[0780] Step 7:

[0781] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model. Calls the generative AI model's API and sends a request to generate an appropriate answer.

[0782] Step 8:

[0783] Generative AI model: Based on the analysis results and emotion recognition results, it generates an appropriate answer to the question, "Please tell me how to issue a resident registration card." For example, it generates an answer such as, "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[0784] Step 9:

[0785] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[0786] Step 10:

[0787] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[0788] Step 11:

[0789] Device: The user enters their evaluation and opinion on the provided answer and presses the send button, sending a feedback message such as "The answer was easy to understand" or "I'd like to know more."

[0790] Step 12:

[0791] Server: Receives feedback sent from the device and stores it in a database. Passes the received feedback to the generative AI model and emotion engine, and uses it to improve future inquiry responses and response accuracy.

[0792] Example 2

[0793] 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."

[0794] Conventional automated response systems had the ability to provide appropriate answers to user inquiries, but they were unable to respond in a way that took the user's emotions into consideration. This made improving user satisfaction a challenge. Since the system lacked appropriate responses, particularly for users who were confused or angry, there was a need to further improve the quality of response to inquiries.

[0795] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0796] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the content of the received inquiry, means for using a natural language processing engine to analyze the content of the inquiry, means for recognizing the user's emotions, means for generating an appropriate answer based on the analysis result and the emotion recognition result, means for providing the generated answer to the user, means for receiving feedback on the provided answer, and means for using the received feedback and emotion data as learning data. This enables appropriate responses and answers that take the user's emotions into consideration, improving the quality of inquiry responses and increasing user satisfaction.

[0797] "Means for receiving inquiries from users" refers to a function that allows the server to receive the content of inquiries entered by users through the government's website or application.

[0798] The "means for analyzing the content of the received inquiry" is a function that analyzes the content of the user's inquiry received by the server in order to understand it.

[0799] "Means for using a natural language processing engine" refers to a function that uses a software engine that performs natural language processing when analyzing the content of a received inquiry.

[0800] The "means for recognizing the user's emotions" is a function for identifying the emotional state at the time based on the content of the inquiry sent by the user.

[0801] "Means for generating appropriate answers based on analysis results and emotion recognition results" refers to a function for generating appropriate answers based on the results of natural language processing and emotion recognition.

[0802] "Means for providing the generated answer to the user" refers to the function by which the server sends the answer created by the generative AI model to the user's terminal and displays it.

[0803] The "means for receiving feedback on the provided answer" is a function that allows a user to input opinions and evaluations on the provided answer, and the server receives them.

[0804] The "means for using received feedback and emotion data as learning data" is a function for using the feedback and emotion recognition results received from the user to improve the system's performance and response accuracy.

[0805] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates answers to the inquiries, and recognizes the emotions of the users to provide appropriate responses. Hereinafter, embodiments of the present invention will be described in detail.

[0806] Overall system configuration and functions

[0807] The system consists of the following main components:

[0808] 1. User terminal: A device with an interface that allows users to make inquiries using government websites and applications. This includes PCs, smartphones, tablets, etc.

[0809] 2. Server: A central computer system that receives user queries, analyzes them, and generates answers. The server uses the following software and programs:

[0810] Natural language processing engines (e.g., SpaCy and BERT)

[0811] Emotion engine (e.g., IBM Watson Tone Analyzer)

[0812] Generative AI models (e.g., GPT-3)

[0813] Database (e.g. MySQL)

[0814] 3. Natural language processing engine: Software or a program that analyzes user inquiries, extracts the intent and keywords of the inquiry, and passes them on to the next process.

[0815] 4. Generative AI model: An artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of the natural language processing engine.

[0816] 5. Emotion Engine: Software or a program that recognizes the user's emotions based on the query or user input. The recognized emotions are used to generate responses and adjust the tone.

[0817] 6. Database: A storage system for storing data on queries, generated answers, user feedback, and recognized emotions.

[0818] Program processing flow

[0819] 1. Receiving Inquiries

[0820] Device: The user opens the government website or application, enters "Please tell me how to issue a resident registration card" into the inquiry form, and presses the send button.

[0821] 2. Analysis of inquiry content

[0822] Server: Receives the inquiry sent by the user and temporarily stores it in a local database. The stored inquiry is assigned a unique identification number (e.g., UUID).

[0823] Server: Sends the received query content to a natural language processing engine (e.g., SpaCy or BERT) by sending an HTTP POST request to the API endpoint of the natural language processing engine.

[0824] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[0825] 3. User Emotion Recognition

[0826] Server: Sends the analysis results received from the natural language processing engine to the emotion engine (e.g., IBM Watson Tone Analyzer). Sends an HTTP POST request to the emotion engine's API endpoint.

[0827] Emotion engine: Based on the user's input, it recognizes emotions such as excitement, anger, confusion, and joy and returns them to the server.

[0828] 4. Answer Generation

[0829] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model (e.g., GPT-3).

[0830] Generative AI model: Based on the emotion recognition results, it generates a response with a tone and content that matches the user's emotions. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[0831] 5. Providing generated answers

[0832] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[0833] 6. Displaying the Answer to the User

[0834] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[0835] Specific examples

[0836] For example, a user types, "Please tell me how to issue a resident registration card," and this is sent to the server. The server uses a natural language processing engine to analyze the content and extract keywords. At the same time, the emotion engine recognizes the user's emotion as "confusion." Based on this analysis result and the emotion recognition result, the generative AI model generates a response in a tone that alleviates the "confusion," such as, "We will answer your questions about how to issue a resident registration card. Please come to the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This allows the user to resolve the problem with peace of mind.

[0837] Prompt Sentence Examples

[0838] example:

[0839] User's question: "How do I get a certificate of residence issued?"

[0840] Analysis results: {"keywords": ["resident card", "issuance", "method"], "intent": "information_request"}

[0841] Emotion recognition result: "confused"

[0842] Based on this prompt, a generative AI model generates an appropriate response, and the system as a whole accumulates feedback and sentiment data that can be used to improve future inquiries.

[0843] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0844] Step 1:

[0845] The user opens the city hall's website or application, enters "Please tell me how to issue a resident registration card" into the inquiry form, and presses the send button.

[0846] Input: User's inquiry: "Please tell me how to issue a resident registration card."

[0847] Specific operation: When the user clicks the "Send" button, the terminal sends the input contents to the server as an HTTP POST request.

[0848] Output: The query received by the server.

[0849] Step 2:

[0850] The server receives the query sent by the user and temporarily stores it in a local database, where it is assigned a unique identification number (e.g., UUID).

[0851] Input: User's inquiry: "Please tell me how to issue a resident registration card."

[0852] Specific operation: The server saves the inquiry in the "inquiries" table of the MySQL database and issues a unique ID.

[0853] Output: The query and its unique identification number stored in the database.

[0854] Step 3:

[0855] The server sends the received query to a natural language processing engine.

[0856] Input: The query and its unique identification number stored in the database.

[0857] Specific operation: The server converts the query content into JSON format and sends an HTTP POST request to the API endpoint of a natural language processing engine (e.g., SpaCy or BERT).

[0858] Output: The results of the analysis by the natural language processing engine are received in JSON format.

[0859] Step 4:

[0860] The natural language processing engine analyzes the inquiry, extracts keywords such as "resident registration," "issuance," and "method," and analyzes the sentence structure.

[0861] Input: JSON data of the query sent from the server.

[0862] Specific operation: The natural language processing engine performs text analysis, compiles the extracted keywords and intent into JSON format, and returns it to the server.

[0863] Output: JSON format data containing the parsed results.

[0864] Step 5:

[0865] The server sends the analysis results from the natural language processing engine to the emotion engine.

[0866] Input: JSON data of the analysis results obtained from the natural language processing engine.

[0867] Specific operation: The server sends the analysis results to the API endpoint of the emotion engine (e.g., IBM Watson Tone Analyzer) via an HTTP POST request.

[0868] Output: Receives emotion data recognized by the emotion engine in JSON format.

[0869] Step 6:

[0870] The emotion engine recognizes the user's emotions based on the content of the query and identifies emotions such as "confusion."

[0871] Input: JSON data of the analysis result sent from the server.

[0872] Specific operation: The emotion engine performs emotion analysis and returns the recognized emotion tag to the server.

[0873] Output: JSON formatted data containing emotion recognition results.

[0874] Step 7:

[0875] The server sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model.

[0876] Input: JSON data of analysis results and emotion recognition results obtained from the natural language processing engine and emotion engine.

[0877] Specific operation: The server sends a prompt message to the API endpoint of the generative AI model (e.g., GPT-3) based on the analysis results and emotion recognition results.

[0878] Output: JSON-formatted data containing the answers generated by the generative AI model.

[0879] Step 8:

[0880] The generative AI model generates responses with a tone and content that matches the user's emotions based on the emotion recognition results.

[0881] Input: Prompt sentence based on analysis results and emotion recognition results.

[0882] Specific operation: The generative AI model generates an appropriate answer and returns it to the server in JSON format.

[0883] Output: JSON-formatted data containing the appropriate answer from the generative AI model.

[0884] Step 9:

[0885] The server formats the answers received from the generative AI model and converts them into data for transmission to the user's device.

[0886] Input: JSON data of the appropriate answer data from the generative AI model.

[0887] What happens: The server embeds the response in an HTML template and reformats it for display in the user's browser.

[0888] Output: Formatted response data.

[0889] Step 10:

[0890] The user terminal receives the response data sent from the server and displays it on the user interface.

[0891] Input: Formatted response data.

[0892] Specific operation: The web browser on the user's device renders the received HTML data and displays it to the user.

[0893] Output: Answers displayed on the screen.

[0894] Step 11:

[0895] The user enters feedback on the provided answer and presses the submit button.

[0896] Input: The feedback entered by the user.

[0897] Specific operation: When a user fills out the feedback form and presses the submit button, the terminal sends the feedback to the server as an HTTP POST request.

[0898] Output: The feedback received by the server.

[0899] Step 12:

[0900] The server receives the feedback sent from the device and stores it in a database. The received feedback and emotional data are used as training data for the generative AI model.

[0901] Input: User feedback.

[0902] What it does: The server stores the feedback in a "feedback" table in the database and uses it for subsequent analysis and improvement processes.

[0903] Output: Feedback content and sentiment data stored in a database.

[0904] (Application example 2)

[0905] 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."

[0906] Conventional query processing systems do not recognize the user's emotions and generate answers based on them, but simply respond based on keywords, which can result in users not receiving satisfactory answers. Furthermore, when a user requests recommendations for content such as movies or TV dramas, there is a problem in that appropriate recommendations based on the user's emotions cannot be made.

[0907] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the received inquiry content, means for recognizing the user's emotions based on the analyzed content, means for generating an appropriate answer according to the user's emotions, means for providing the generated answer to the user, means for receiving feedback on the provided answer, means for using the received feedback as learning data, and means for recommending movies and TV shows. This makes it possible to respond to inquiries and recommend appropriate movies and TV shows that take the user's emotions into consideration.

[0908] "Means for receiving inquiries from users" refers to a function for receiving questions or requests entered by users through devices such as smartphones or computers from those devices.

[0909] The "means for analyzing the content of the received inquiry" is a function for understanding the content of the inquiry received from the user and performing processing to extract appropriate information.

[0910] The "means for recognizing the user's emotions based on the analyzed content" is a function for identifying the emotions the user is feeling based on the analysis results of the inquiry content.

[0911] The "means for generating an appropriate response according to the user's emotions" is a function that takes into account the recognized emotions of the user and creates an answer with a tone and content that is appropriate to those emotions.

[0912] The "means for providing the generated answer to the user" is a function for displaying or replying to the generated answer on the user's device.

[0913] The "means for receiving feedback on the provided answer" is a function that allows a user to input and receive an evaluation or additional comments on the provided answer.

[0914] "Means of using received feedback as learning data" is a function that accumulates feedback from users and uses it to improve the accuracy of future inquiries and to improve the system.

[0915] The "means for recommending movies and TV dramas" is a function for recommending appropriate movies and TV dramas to users based on the user's emotions and inquiries.

[0916] MODE FOR CARRYING OUT THE INVENTION

[0917] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates responses to the inquiries, and further recognizes the emotions of the users to provide appropriate responses. Hereinafter, specific embodiments of the present invention will be described.

[0918] Overall system configuration and functions

[0919] The system consists of the following main components:

[0920] 1. User Device:

[0921] It is a device with an interface that allows users to make inquiries through a smartphone application.

[0922] Hardware: Smartphone (e.g. iPhone, Android device)

[0923] Software: User Interface Application

[0924] 2. Server:

[0925] It is a central computer system that receives and processes user queries, analyzes the query, generates answers, provides feedback, and recognizes emotions.

[0926] Hardware: Server computer

[0927] Software: Web server, database server

[0928] 3. Natural Language Processing Engine:

[0929] This software analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to the next process.

[0930] Usage examples: SpaCy, NLTK

[0931] 4. Generative AI Models:

[0932] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[0933] Usage example: GPT-4, BERT

[0934] 5. Emotion Engine:

[0935] This software recognizes user emotions based on the content of the query and user input, and uses the recognized emotions to generate responses and adjust tone.

[0936] Usage examples: IBM Watson Tone Analyzer, Microsoft Azure Text Analytics

[0937] 6. Database:

[0938] It is a storage system for storing data on inquiries, generated answers, user feedback, and recognized emotions.

[0939] Usage example: MySQL, PostgreSQL

[0940] 7. Movie and TV series recommendation module:

[0941] This software recommends appropriate movies and dramas based on the user's emotions and inquiries.

[0942] Program processing

[0943] This program operates as follows:

[0944] 1. Receiving Inquiries

[0945] The user terminal receives an inquiry from the user through a smartphone app, which is then sent to the server.

[0946] 2. Analysis of inquiry content

[0947] The server sends the received inquiry to a natural language processing engine, which analyzes the inquiry and extracts keywords.

[0948] 3. User Emotion Recognition

[0949] The server receives the analysis results from the natural language processing engine and sends them to the emotion engine, which recognizes the user's emotions and returns the results to the server.

[0950] 4. Answer Generation

[0951] The server sends the analysis results and emotion recognition results to a generative AI model, which generates an answer appropriate to the user's emotions. For example, if the system recognizes that the user is "depressed," it will "recommend movies and TV shows that will help alleviate the depression."

[0952] Prompt Sentence Examples

[0953] If the user is in an emotionally depressed state, recommend movies or TV shows based on the following: "I'm feeling down. Can you recommend some movies that will cheer me up?"

[0954] 5. Providing generated answers

[0955] The server formats the generated answer, converts it into data for display on the user's device, and transmits it to the user. The user can view the answer through the user interface on their smartphone.

[0956] 6. Receive feedback and learn

[0957] The user terminal receives feedback from the user regarding the provided answer and sends it to the server, which stores this feedback in a database and uses it to improve future responses to inquiries.

[0958] In this way, the present invention can recognize the user's emotions and generate appropriate responses, thereby making it possible to improve the efficiency of inquiry response work and user satisfaction.

[0959] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0960] Step 1:

[0961] The user uses a smartphone app to input the inquiry and press the send button. An example input is "I'm feeling down, can you recommend a movie that will cheer me up?" The device then sends this inquiry to the server. The input data is the inquiry in text format, and the output is the data transferred to the server.

[0962] Step 2:

[0963] The server sends the received query content to a natural language processing engine, which analyzes the content and extracts key keywords (e.g., "feeling depressed," "cheerful," "movies"). The input data is the user's query content, and the output data is the analysis results including the extracted keywords.

[0964] Step 3:

[0965] The server sends the analysis results received from the natural language processing engine to the emotion engine. The emotion engine recognizes the user's emotions based on these analysis results. For example, it identifies emotions such as "feeling depressed." The input data is the analysis results, and the output data is the recognized emotional information.

[0966] Step 4:

[0967] The server sends the analysis results and emotion recognition results to the generative AI model. The generative AI model generates an answer that is appropriate to the user's emotions based on these inputs. For example, if the emotion is recognized as "depressed," it generates a recommendation such as "comedy movies" as an "uplifting movie." The input data are the analysis results and emotion recognition results, and the output data is the generated answer. The specific operation uses the following prompt sentence:

[0968] If the user is in an emotionally depressed state, recommend movies or TV shows based on the following: "I'm feeling down. Can you recommend some movies that will cheer me up?"

[0969] Step 5:

[0970] The server formats the generated answer and converts it into an appropriate data format (e.g., JSON, XML, HTML) for transmission to the device. The input data is the generated answer, and the output data is the formatted answer.

[0971] Step 6:

[0972] The terminal receives the formatted answer data sent from the server and displays it on the user interface. The user can check recommendations such as "Title A," which is a comedy movie, on the screen. The input data is the formatted answer data, and the output data is the displayed answer.

[0973] Step 7:

[0974] The user inputs and transmits feedback on the provided answer through the terminal. For example, the feedback may include "The recommendation for this movie was accurate." The terminal transmits this feedback to the server. The input data is the user's feedback, and the output data is the feedback transmitted to the server.

[0975] Step 8:

[0976] The server stores the feedback received from the device in a database and associates it with the analysis results and emotion recognition results. This feedback is used to improve future inquiries. The input data is the user feedback, and the output data is the stored feedback information.

[0977] 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.

[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 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.

[0980] [Third embodiment]

[0981] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0982] 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.

[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 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.

[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] 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.

[0989] 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.

[0990] 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.

[0991] 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.

[0992] 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."

[0993] The present invention relates to an automatic response system that accepts inquiries from users, analyzes them, and generates and provides appropriate answers, thereby improving the efficiency of government office operations and improving public services. The following describes specific embodiments of the present invention.

[0994] Overall system configuration

[0995] The system consists of the following main components:

[0996] 1. User Device

[0997] It is a device that has an interface that allows citizens to make inquiries using government websites or applications.

[0998] 2. Server

[0999] A central computer system that receives and processes user inquiries, including analyzing the inquiry, generating answers, and receiving feedback.

[1000] 3. Natural Language Processing Engine

[1001] This is software or a program that analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to subsequent processing.

[1002] 4. Generative AI Models

[1003] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[1004] 5. Database

[1005] It is a storage system for saving data such as inquiry details, generated answers, and feedback from users.

[1006] Program processing

[1007] The program processing of this system will be explained in natural language below.

[1008] 1. Receiving Inquiries

[1009] Terminal: The user enters an inquiry about how to issue a resident registration card, etc., through the government website or application and presses the send button.

[1010] Server: Receives the inquiry sent from the terminal and temporarily stores it in a database.

[1011] 2. Analysis of inquiry content

[1012] Server: Sends the received query content to the natural language processing engine.

[1013] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident registration card," "issuance," and "method," and clarifies the subject's intent.

[1014] Server: Passes the analysis results to the generative AI model.

[1015] 3. Answer Generation

[1016] Generative AI model: Generates appropriate answers to user inquiries based on the analysis results of the natural language processing engine.

[1017] Generative AI model: Creates a response that includes detailed information and escalation as needed. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID."

[1018] 4. Providing generated answers

[1019] Server: Formats the answers received from the generative AI model and sends them to the device.

[1020] Terminal: Displays the generated answer to the user.

[1021] 5. Receiving Feedback

[1022] Terminal: The user enters feedback on the provided answer and presses the submit button.

[1023] Server: Receives feedback sent from devices and stores it in a database.

[1024] 6. Use of Feedback

[1025] Server: Passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy.

[1026] Specific examples

[1027] For example, when a user types "Please tell me how to issue a resident registration card" into a device and sends it, the server receives this and requests analysis from the natural language processing engine. Keywords such as "resident registration card," "issuance," and "method" are extracted as a result of the analysis. Based on this, the generative AI model generates a response such as "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and sends this to the user's device via the server.

[1028] In this way, the automatic response system based on the present invention can improve the efficiency of inquiries handled by government offices and the quality of services to citizens.

[1029] The processing flow will be explained below.

[1030] Step 1:

[1031] Terminal: The user enters an inquiry on the government website or application and presses the send button. For example, they might enter, "Please tell me how to issue a resident registration card."

[1032] Step 2:

[1033] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[1034] Step 3:

[1035] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[1036] Step 4:

[1037] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[1038] Step 5:

[1039] Server: Sends the analysis results received from the natural language processing engine to the generative AI model. Passes the analysis results to the generative AI model's API as input data.

[1040] Step 6:

[1041] Generative AI model: Generates an appropriate response based on the analysis results. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[1042] Step 7:

[1043] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[1044] Step 8:

[1045] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[1046] Step 9:

[1047] Terminal: The user enters feedback on the provided answer and presses the submit button. Feedback such as "The answer was easy to understand" or "I would like more specific information" is sent.

[1048] Step 10:

[1049] Server: Receives feedback sent from the device and stores it in a database. Passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy.

[1050] Example 1

[1051] 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."

[1052] Conventional government office inquiry response systems have struggled to provide fast and accurate responses to large volumes of inquiries, resulting in labor shortages and increased workloads. They also lacked a system for efficiently utilizing feedback on the answers provided to improve the accuracy of the system. Furthermore, while there is a need to respond to user inquiries in real time, there are also technical challenges in achieving this.

[1053] 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.

[1054] In this invention, the server includes means for receiving inquiries from users, means for saving the received inquiry content in a database, means for analyzing the inquiry content using a natural language processing engine, means for generating an answer using a generative AI model based on the analysis result, means for providing the generated answer to the user terminal, means for receiving user feedback and saving it in a database, and means for using the received feedback as learning data for the generative AI model to improve the accuracy of the answer, thereby enabling improved efficiency and accuracy in responding to inquiries and real-time user responses.

[1055] "Users" refers to citizens and residents who use the government's inquiry system.

[1056] "Inquiries" refer to questions or requests that users enter and submit through government websites or applications.

[1057] "Terminal" refers to a device used by a user to enter and send a query.

[1058] "Server" refers to a central computer system that receives and processes user queries.

[1059] A "natural language processing engine" refers to software or a program for analyzing the content of user inquiries.

[1060] A "generative AI model" refers to an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[1061] "Answer" refers to the appropriate response generated by the generative AI model based on the user's inquiry.

[1062] "Feedback" refers to the evaluation or opinion that a user enters in response to a provided answer.

[1063] A "database" refers to a storage system for storing data such as inquiries, generated responses, and user feedback.

[1064] "Training data" refers to data used to improve the performance of a generative AI model based on received feedback.

[1065] The present invention relates to an automatic response system that accepts inquiries from users, analyzes them, and generates and provides appropriate answers, thereby improving the efficiency of government office operations and improving public services. The following describes specific embodiments of the present invention.

[1066] Overall system configuration

[1067] The system consists of the following main components:

[1068] 1. User Device

[1069] This is a device with an interface that allows citizens to make inquiries using government websites or applications. This device can be any common computing device such as a personal computer (PC), smartphone, or tablet.

[1070] 2. Server

[1071] A central computer system that receives and processes user inquiries, analyzes the content of the inquiry, generates answers, and receives feedback. This server can be a cloud-based server or an on-premise server.

[1072] 3. Natural Language Processing Engine

[1073] This is software and a program for analyzing user inquiries. This engine analyzes the main keywords and intent of the inquiry and passes it on to subsequent processing. Specific software used is a Python natural language processing library (e.g., NLTK or spaCy).

[1074] 4. Generative AI Models

[1075] This is an artificial intelligence model that generates appropriate answers based on the analysis results of a natural language processing engine. This AI model is built using a machine learning framework (e.g., TensorFlow or PyTorch). The generative AI model continuously learns based on user feedback to improve the accuracy of its answers.

[1076] 5. Database

[1077] A storage system for saving data such as inquiry contents, generated answers, and user feedback, etc. It uses a relational database (e.g., MySQL, PostgreSQL) or a NoSQL database (e.g., MongoDB, DynamoDB).

[1078] Program processing

[1079] The program processing of this system is carried out in the following manner.

[1080] 1. Receiving Inquiries

[1081] Terminal: The user enters a question, such as how to issue a resident registration card, through a website or application and presses the send button. For example, the user enters a question such as "How do I issue a resident registration card?"

[1082] Server: Receives the inquiry sent from the user terminal and temporarily stores it in a database.

[1083] 2. Analysis of inquiry content

[1084] Server: Sends the received query content to the natural language processing engine.

[1085] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident registration card," "issuance," and "method," and returns the analysis results to the server.

[1086] 3. Answer Generation

[1087] Server: Passes the analysis results from the natural language processing engine to the generative AI model.

[1088] Generative AI model: Based on the analysis results, it generates a specific answer such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is identification."

[1089] 4. Providing generated answers

[1090] Server: Sends the answer received from the generative AI model to the user's device.

[1091] Terminal: Displays the generated answer to the user, for example, as text on a web page or in the form of a chatbot within an application.

[1092] 5. Receiving Feedback

[1093] Terminal: The user enters feedback on the provided answer and presses the submit button.

[1094] Server: Receives feedback sent from devices and stores it in a database.

[1095] 6. Use of Feedback

[1096] Server: The received feedback is used as learning data for the generative AI model. Based on this feedback data, the generative AI model re-learns and aims to improve the accuracy of its answers.

[1097] Specific examples

[1098] For example, suppose a user types and sends a query on their device, "Please tell me how to issue a resident registration card." The server receives this query and requests analysis from a natural language processing engine. Keywords such as "resident registration card," "issuance," and "method" are extracted as a result of the analysis. Based on this, the generative AI model generates a response: "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This response is sent to the user's device via the server. After seeing this response, the user enters feedback, which is received by the server and stored in a database. The generative AI model then re-learns based on the feedback, allowing it to provide more accurate responses to future inquiries.

[1099] Prompt Sentence Examples

[1100] "A user has submitted the inquiry '[How do I issue a residence certificate?]'. Please generate an appropriate answer based on this information."

[1101] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1102] Step 1:

[1103] Receiving inquiries

[1104] Terminal: The user enters the details of their inquiry (e.g., "Please tell me how to issue a resident registration card") through the government's website or application and presses the send button.

[1105] Input: Text data of the inquiry.

[1106] Server: Receives inquiries sent from user terminals and temporarily stores them in a database.

[1107] Output: The query is saved in the database.

[1108] Step 2:

[1109] Analysis of inquiry content

[1110] Server: Sends the received query content to the natural language processing engine.

[1111] Input: Text data of the inquiry content stored in the database.

[1112] Natural language processing engine: Analyzes the content of the inquiry and extracts keywords such as "resident certificate," "issuance," and "method." The analysis results are returned to the server.

[1113] Output: Analysis results including extracted keywords and query intent.

[1114] Step 3:

[1115] Generate answers

[1116] Server: Passes the analysis results from the natural language processing engine to the generative AI model.

[1117] Input: Analysis result data.

[1118] Generative AI model: Based on the analysis results, it generates appropriate answers to user inquiries. It generates specific answers such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID."

[1119] Output: The text data of the generated answers.

[1120] Step 4:

[1121] Providing generated answers

[1122] Server: Formats the answers received from the generative AI model and sends them to the user's device. To make the answers easier to view, the server formats the data in HTML or JSON format.

[1123] Input: Text data of the answer from the generative AI model.

[1124] Terminal: The generated answer is displayed to the user. For example, the answer can be displayed on a web page or in the form of a chatbot within an application.

[1125] Output: The answer displayed to the user.

[1126] Step 5:

[1127] Receiving Feedback

[1128] Terminal: The user enters feedback (e.g., "helpful" or "not helpful") for the provided answer and presses the submit button. It is also possible to enter specific comments.

[1129] Input: User feedback data.

[1130] Server: Receives feedback sent from devices and stores it in a database.

[1131] Output: Feedback data stored in a database.

[1132] Step 6:

[1133] Using Feedback

[1134] Server: Passes received feedback to the generative AI model.

[1135] Input: Feedback data stored in the database.

[1136] Generative AI model: The model is retrained based on the received feedback data to improve the accuracy of answers and the quality of responses.

[1137] Output: The retrained generative AI model.

[1138] The above is the specific program processing flow of the system, which improves the efficiency and accuracy of inquiries and enables real-time user support.

[1139] (Application example 1)

[1140] 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."

[1141] In modern electronic payment services, it is important to respond quickly and accurately to user inquiries. However, conventional systems take a long time to analyze the inquiry content and generate answers, and the answers provided are often not accurate enough. This makes it difficult to improve user satisfaction. Another problem is that feedback is not fully utilized, hindering system improvement. To solve these issues, there is a need for the development of a system that provides quick and accurate answers and efficiently utilizes feedback as training data.

[1142] 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.

[1143] In this invention, the server includes means for receiving inquiries from users, means for analyzing the received inquiries, means for generating appropriate answers based on the inquiries, means for providing the generated answers to the users, means for receiving feedback on the provided answers, means for using the received feedback as training data, means for identifying inquiries related to payments, and means for generating payment-related information. This enables quick and accurate analysis of inquiries and generation of answers in electronic payment services, thereby improving user satisfaction. Furthermore, by efficiently utilizing feedback as training data, the system can be continuously improved.

[1144] A "user inquiry" is the act of using an electronic device to ask a question or request information or a service.

[1145] "Means for receiving" refers to the functions and methods for obtaining information via electronic devices or communication systems.

[1146] "Means for analyzing the content of inquiries" refers to techniques and methods for understanding the meaning and intent of the information obtained and extracting relevant information.

[1147] "Means for generating appropriate answers" refers to techniques and methods for providing accurate and useful information to users' questions based on the analysis results.

[1148] "Means for providing the generated answer to the user" refers to the method or technology for displaying the generated answer on the user's terminal.

[1149] "Means for receiving feedback" refers to techniques and methods for receiving evaluations and opinions from users.

[1150] "Means of using feedback as training data" refers to techniques and methods for improving system performance based on received evaluations and opinions.

[1151] "Payment Inquiry" means a question or request from a user regarding the payment procedure or process.

[1152] "Means for generating payment information" refers to the technology or method for organizing and providing information related to payment procedures.

[1153] The present invention relates to an automatic response system that accepts and analyzes inquiries from users, and generates and provides appropriate responses, thereby improving the efficiency of electronic payment services and increasing user satisfaction. The following describes specific embodiments of the present invention.

[1154] Overall system configuration

[1155] The system consists of the following main components:

[1156] 1. User Device

[1157] A device with an interface that allows users to make inquiries using an electronic payment service application. Typical examples are smartphones and tablets.

[1158] 2. Server

[1159] A central computer system that receives inquiries, analyzes them, generates answers, and receives feedback. Often runs on a cloud-based server.

[1160] 3. Natural Language Processing Engine

[1161] This is software that analyzes the content of user inquiries and extracts key keywords and intent. Here, we use SpaCy (ja_core_news_sm model).

[1162] 4. Generative AI Models

[1163] This is an AI model that generates appropriate answers based on the analysis results of a natural language processing engine. It uses Huggingface's Transformers library.

[1164] 5. Database

[1165] It is a storage system for saving data such as inquiries, generated answers, and feedback from users.

[1166] System Program Processing

[1167] The server first receives the inquiry sent from the user's device. Next, it sends that content to a natural language processing engine (SpaCy) for analysis. The keywords and intent extracted as a result of the analysis are passed to a generative AI model (Transformers), which generates an appropriate answer. The generated answer is sent back to the user's device and provided to the user. When the user sends feedback on the answer provided, the server receives that feedback and stores it in a database. The received feedback is used as learning data for the generative AI model.

[1168] Specific examples

[1169] For example, if a user types "What should I do if my credit card payment doesn't work?" into a smartphone app and submits it, this inquiry is sent to the server. On the server, SpaCy's natural language processing engine extracts keywords such as "credit card," "payment," "doesn't work," and "how to fix it." Based on the analysis results, the generative AI model generates an answer such as "If your credit card payment doesn't work, please double-check your card information and try again. If the problem persists, please contact customer support," and provides it to the user.

[1170] Prompt Sentence Examples

[1171] What should I do if my credit card payment doesn't work?

[1172] The above is a specific embodiment for carrying out the present invention. This system enables quick and accurate analysis of user inquiries in electronic payment services and provides appropriate answers. Furthermore, by efficiently utilizing feedback as learning data, the system can be continuously improved.

[1173] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1174] Step 1:

[1175] User enters a query

[1176] A user uses a smartphone app for an electronic payment service to enter an inquiry and press the send button. As input, the user enters the question, "What should I do if my credit card payment doesn't go through?". As output, this inquiry is sent to the server.

[1177] Step 2:

[1178] The server receives the query

[1179] The server receives the inquiry sent from the user terminal. The input is text data (question) from the user. The output is to temporarily store the received inquiry in a database.

[1180] Step 3:

[1181] The server sends the query to the natural language processing engine

[1182] The server sends the received query to a natural language processing engine (SpaCy) for analysis. The query stored on the server is used as input. The analysis results (extracted keywords and intent) are obtained as output.

[1183] Step 4:

[1184] The natural language processing engine performs the analysis.

[1185] A natural language processing engine analyzes the content of an inquiry and extracts key keywords and intent. The input is text data (the question). The output is keywords such as "credit card," "payment," "not working," and "solution."

[1186] Step 5:

[1187] The server sends the analysis results to the generated AI model

[1188] The server sends the analysis results of the natural language processing engine to the generative AI model (Transformers). The input is the analysis results (extracted keywords and intent). The output is the generated answer.

[1189] Step 6:

[1190] Generative AI models generate answers

[1191] The generative AI model generates an appropriate answer based on the analysis results. The input is the analysis result, such as "My credit card payment is not working" and "What should I do?" The output is the answer, "If your credit card payment is not working, please double-check your card information and try again. If the problem persists, please contact customer support."

[1192] Step 7:

[1193] The server provides the generated answer to the user

[1194] The server formats the answer received from the generative AI model and sends it to the user's device. The input is the generated answer. The output is the answer displayed on the user's device.

[1195] Step 8:

[1196] Users can provide feedback on the answers provided

[1197] The user inputs feedback for the provided answer and presses the submit button. The input is the feedback content. The output is the feedback sent to the server.

[1198] Step 9:

[1199] Server receives feedback

[1200] The server receives the feedback sent from the user terminal and stores it in a database. As input, there is feedback data from the user. As output, the received feedback is stored in the database.

[1201] Step 10:

[1202] The server sends the feedback as training data to the generative AI model

[1203] The server passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy. The input is the received feedback data. The output is an attempt to improve the performance of the generative AI model.

[1204] 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.

[1205] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates answers to the inquiries, and recognizes the emotions of the users to provide appropriate responses. Hereinafter, embodiments of the present invention will be described in detail.

[1206] Overall system configuration and functions

[1207] The system consists of the following main components:

[1208] 1. User Device

[1209] It is a device with an interface that allows users to make inquiries using government websites and applications.

[1210] 2. Server

[1211] It is a central computer system that receives and processes user queries, analyzes the query, generates answers, provides feedback, and recognizes emotions.

[1212] 3. Natural Language Processing Engine

[1213] This is software or a program that analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to the next process.

[1214] 4. Generative AI Models

[1215] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[1216] 5. Emotion Engine

[1217] It is software or a program that recognizes a user's emotions based on the content of the query or user input, and uses the recognized emotions to generate a response or adjust the tone.

[1218] 6. Database

[1219] It is a storage system for storing data on inquiries, generated answers, user feedback, and recognized emotions.

[1220] Program processing

[1221] The program processing of this system will be explained in natural language below.

[1222] 1. Receiving Inquiries

[1223] Terminal: The user enters an inquiry about how to issue a resident registration card, etc., through the government website or application and presses the send button.

[1224] 2. Analysis of inquiry content

[1225] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[1226] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[1227] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[1228] 3. User Emotion Recognition

[1229] Server: Sends the analysis results received from the natural language processing engine to the emotion engine. Calls the emotion engine's API and sends a request to recognize the user's emotion.

[1230] Emotion engine: Based on the user's input, it recognizes emotions such as excitement, anger, confusion, and joy and returns them to the server.

[1231] 4. Answer Generation

[1232] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model.

[1233] Generative AI model: Based on the emotion recognition results, it generates a response with a tone and content that matches the user's emotions. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[1234] 5. Providing generated answers

[1235] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[1236] 6. Displaying the Answer to the User

[1237] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[1238] 7. Receiving Feedback

[1239] Terminal: The user enters feedback on the provided answer and presses the submit button. Feedback such as "The answer was easy to understand" or "I would like more specific information" is sent.

[1240] 8. Use of Feedback and Emotional Data

[1241] Server: Receives feedback sent from the device and stores it in a database. The received feedback and emotional data are passed to the generative AI model, which uses it to improve future response and response accuracy.

[1242] Specific examples

[1243] For example, a user types, "Please tell me how to issue a resident registration card," and this is sent to the server. The server analyzes the content using a natural language processing engine and extracts keywords. At the same time, the emotion engine recognizes the user's emotion as "confusion." Based on the analysis results and emotion recognition results, the generative AI model generates a response in a tone that alleviates the "confusion," such as, "We will answer your questions about how to issue a resident registration card. Please come to the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This allows the user to resolve the problem with peace of mind.

[1244] In this way, the present invention can recognize the user's emotions and generate appropriate responses, thereby making it possible to improve the efficiency of government office inquiries and further enhance the quality of public services.

[1245] The processing flow will be explained below.

[1246] Step 1:

[1247] Terminal: The user uses the government's website or application to enter an inquiry such as "Please tell me how to issue a resident registration card" and presses the send button.

[1248] Step 2:

[1249] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[1250] Step 3:

[1251] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[1252] Step 4:

[1253] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[1254] Step 5:

[1255] Server: Sends the analysis results received from the natural language processing engine to the emotion engine. Calls the emotion engine's API and sends a request to recognize the user's emotion.

[1256] Step 6:

[1257] Emotion engine: Based on the content of the inquiry, it recognizes the user's emotions such as "confusion," "anger," and "joy," and returns the emotional data to the server.

[1258] Step 7:

[1259] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model. Calls the generative AI model's API and sends a request to generate an appropriate answer.

[1260] Step 8:

[1261] Generative AI model: Based on the analysis results and emotion recognition results, it generates an appropriate answer to the question, "Please tell me how to issue a resident registration card." For example, it generates an answer such as, "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[1262] Step 9:

[1263] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[1264] Step 10:

[1265] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[1266] Step 11:

[1267] Device: The user enters their evaluation and opinion on the provided answer and presses the send button, sending a feedback message such as "The answer was easy to understand" or "I'd like to know more."

[1268] Step 12:

[1269] Server: Receives feedback sent from the device and stores it in a database. Passes the received feedback to the generative AI model and emotion engine, and uses it to improve future inquiry responses and response accuracy.

[1270] Example 2

[1271] 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."

[1272] Conventional automated response systems had the ability to provide appropriate answers to user inquiries, but they were unable to respond in a way that took the user's emotions into consideration. This made improving user satisfaction a challenge. Since the system lacked appropriate responses, particularly for users who were confused or angry, there was a need to further improve the quality of response to inquiries.

[1273] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1274] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the content of the received inquiry, means for using a natural language processing engine to analyze the content of the inquiry, means for recognizing the user's emotions, means for generating an appropriate answer based on the analysis result and the emotion recognition result, means for providing the generated answer to the user, means for receiving feedback on the provided answer, and means for using the received feedback and emotion data as learning data. This enables appropriate responses and answers that take the user's emotions into consideration, improving the quality of inquiry responses and increasing user satisfaction.

[1275] "Means for receiving inquiries from users" refers to a function that allows the server to receive the content of inquiries entered by users through the government's website or application.

[1276] The "means for analyzing the content of the received inquiry" is a function that analyzes the content of the user's inquiry received by the server in order to understand it.

[1277] "Means for using a natural language processing engine" refers to a function that uses a software engine that performs natural language processing when analyzing the content of a received inquiry.

[1278] The "means for recognizing the user's emotions" is a function for identifying the emotional state at the time based on the content of the inquiry sent by the user.

[1279] "Means for generating appropriate answers based on analysis results and emotion recognition results" refers to a function for generating appropriate answers based on the results of natural language processing and emotion recognition.

[1280] "Means for providing the generated answer to the user" refers to the function by which the server sends the answer created by the generative AI model to the user's terminal and displays it.

[1281] The "means for receiving feedback on the provided answer" is a function that allows a user to input opinions and evaluations on the provided answer, and the server receives them.

[1282] The "means for using received feedback and emotion data as learning data" is a function for using the feedback and emotion recognition results received from the user to improve the system's performance and response accuracy.

[1283] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates answers to the inquiries, and recognizes the emotions of the users to provide appropriate responses. Hereinafter, embodiments of the present invention will be described in detail.

[1284] Overall system configuration and functions

[1285] The system consists of the following main components:

[1286] 1. User terminal: A device with an interface that allows users to make inquiries using government websites and applications. This includes PCs, smartphones, tablets, etc.

[1287] 2. Server: A central computer system that receives user queries, analyzes them, and generates answers. The server uses the following software and programs:

[1288] Natural language processing engines (e.g., SpaCy and BERT)

[1289] Emotion engine (e.g., IBM Watson Tone Analyzer)

[1290] Generative AI models (e.g., GPT-3)

[1291] Database (e.g. MySQL)

[1292] 3. Natural language processing engine: Software or a program that analyzes user inquiries, extracts the intent and keywords of the inquiry, and passes them on to the next process.

[1293] 4. Generative AI model: An artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of the natural language processing engine.

[1294] 5. Emotion Engine: Software or a program that recognizes the user's emotions based on the query or user input. The recognized emotions are used to generate responses and adjust the tone.

[1295] 6. Database: A storage system for storing data on queries, generated answers, user feedback, and recognized emotions.

[1296] Program processing flow

[1297] 1. Receiving Inquiries

[1298] Device: The user opens the government website or application, enters "Please tell me how to issue a resident registration card" into the inquiry form, and presses the send button.

[1299] 2. Analysis of inquiry content

[1300] Server: Receives the inquiry sent by the user and temporarily stores it in a local database. The stored inquiry is assigned a unique identification number (e.g., UUID).

[1301] Server: Sends the received query content to a natural language processing engine (e.g., SpaCy or BERT) by sending an HTTP POST request to the API endpoint of the natural language processing engine.

[1302] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[1303] 3. User Emotion Recognition

[1304] Server: Sends the analysis results received from the natural language processing engine to the emotion engine (e.g., IBM Watson Tone Analyzer). Sends an HTTP POST request to the emotion engine's API endpoint.

[1305] Emotion engine: Based on the user's input, it recognizes emotions such as excitement, anger, confusion, and joy and returns them to the server.

[1306] 4. Answer Generation

[1307] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model (e.g., GPT-3).

[1308] Generative AI model: Based on the emotion recognition results, it generates a response with a tone and content that matches the user's emotions. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[1309] 5. Providing generated answers

[1310] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[1311] 6. Displaying the Answer to the User

[1312] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[1313] Specific examples

[1314] For example, a user types, "Please tell me how to issue a resident registration card," and this is sent to the server. The server uses a natural language processing engine to analyze the content and extract keywords. At the same time, the emotion engine recognizes the user's emotion as "confusion." Based on this analysis result and the emotion recognition result, the generative AI model generates a response in a tone that alleviates the "confusion," such as, "We will answer your questions about how to issue a resident registration card. Please come to the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This allows the user to resolve the problem with peace of mind.

[1315] Prompt Sentence Examples

[1316] example:

[1317] User's question: "How do I get a certificate of residence issued?"

[1318] Analysis results: {"keywords": ["resident card", "issuance", "method"], "intent": "information_request"}

[1319] Emotion recognition result: "confused"

[1320] Based on this prompt, a generative AI model generates an appropriate response, and the system as a whole accumulates feedback and sentiment data that can be used to improve future inquiries.

[1321] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1322] Step 1:

[1323] The user opens the city hall's website or application, enters "Please tell me how to issue a resident registration card" into the inquiry form, and presses the send button.

[1324] Input: User's inquiry: "Please tell me how to issue a resident registration card."

[1325] Specific operation: When the user clicks the "Send" button, the terminal sends the input contents to the server as an HTTP POST request.

[1326] Output: The query received by the server.

[1327] Step 2:

[1328] The server receives the query sent by the user and temporarily stores it in a local database, where it is assigned a unique identification number (e.g., UUID).

[1329] Input: User's inquiry: "Please tell me how to issue a resident registration card."

[1330] Specific operation: The server saves the inquiry in the "inquiries" table of the MySQL database and issues a unique ID.

[1331] Output: The query and its unique identification number stored in the database.

[1332] Step 3:

[1333] The server sends the received query to a natural language processing engine.

[1334] Input: The query and its unique identification number stored in the database.

[1335] Specific operation: The server converts the query content into JSON format and sends an HTTP POST request to the API endpoint of a natural language processing engine (e.g., SpaCy or BERT).

[1336] Output: The results of the analysis by the natural language processing engine are received in JSON format.

[1337] Step 4:

[1338] The natural language processing engine analyzes the inquiry, extracts keywords such as "resident registration," "issuance," and "method," and analyzes the sentence structure.

[1339] Input: JSON data of the query sent from the server.

[1340] Specific operation: The natural language processing engine performs text analysis, compiles the extracted keywords and intent into JSON format, and returns it to the server.

[1341] Output: JSON format data containing the parsed results.

[1342] Step 5:

[1343] The server sends the analysis results from the natural language processing engine to the emotion engine.

[1344] Input: JSON data of the analysis results obtained from the natural language processing engine.

[1345] Specific operation: The server sends the analysis results to the API endpoint of the emotion engine (e.g., IBM Watson Tone Analyzer) via an HTTP POST request.

[1346] Output: Receives emotion data recognized by the emotion engine in JSON format.

[1347] Step 6:

[1348] The emotion engine recognizes the user's emotions based on the content of the query and identifies emotions such as "confusion."

[1349] Input: JSON data of the analysis result sent from the server.

[1350] Specific operation: The emotion engine performs emotion analysis and returns the recognized emotion tag to the server.

[1351] Output: JSON formatted data containing emotion recognition results.

[1352] Step 7:

[1353] The server sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model.

[1354] Input: JSON data of analysis results and emotion recognition results obtained from the natural language processing engine and emotion engine.

[1355] Specific operation: The server sends a prompt message to the API endpoint of the generative AI model (e.g., GPT-3) based on the analysis results and emotion recognition results.

[1356] Output: JSON-formatted data containing the answers generated by the generative AI model.

[1357] Step 8:

[1358] The generative AI model generates responses with a tone and content that matches the user's emotions based on the emotion recognition results.

[1359] Input: Prompt sentence based on analysis results and emotion recognition results.

[1360] Specific operation: The generative AI model generates an appropriate answer and returns it to the server in JSON format.

[1361] Output: JSON-formatted data containing the appropriate answer from the generative AI model.

[1362] Step 9:

[1363] The server formats the answers received from the generative AI model and converts them into data for transmission to the user's device.

[1364] Input: JSON data of the appropriate answer data from the generative AI model.

[1365] What happens: The server embeds the response in an HTML template and reformats it for display in the user's browser.

[1366] Output: Formatted response data.

[1367] Step 10:

[1368] The user terminal receives the response data sent from the server and displays it on the user interface.

[1369] Input: Formatted response data.

[1370] Specific operation: The web browser on the user's device renders the received HTML data and displays it to the user.

[1371] Output: Answers displayed on the screen.

[1372] Step 11:

[1373] The user enters feedback on the provided answer and presses the submit button.

[1374] Input: The feedback entered by the user.

[1375] Specific operation: When a user fills out the feedback form and presses the submit button, the terminal sends the feedback to the server as an HTTP POST request.

[1376] Output: The feedback received by the server.

[1377] Step 12:

[1378] The server receives the feedback sent from the device and stores it in a database. The received feedback and emotional data are used as training data for the generative AI model.

[1379] Input: User feedback.

[1380] What it does: The server stores the feedback in a "feedback" table in the database and uses it for subsequent analysis and improvement processes.

[1381] Output: Feedback content and sentiment data stored in a database.

[1382] (Application example 2)

[1383] 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."

[1384] Conventional query processing systems do not recognize the user's emotions and generate answers based on them, but simply respond based on keywords, which can result in users not receiving satisfactory answers. Furthermore, when a user requests recommendations for content such as movies or TV dramas, there is a problem in that appropriate recommendations based on the user's emotions cannot be made.

[1385] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the received inquiry content, means for recognizing the user's emotions based on the analyzed content, means for generating an appropriate answer according to the user's emotions, means for providing the generated answer to the user, means for receiving feedback on the provided answer, means for using the received feedback as learning data, and means for recommending movies and TV shows. This makes it possible to respond to inquiries and recommend appropriate movies and TV shows that take the user's emotions into consideration.

[1386] "Means for receiving inquiries from users" refers to a function for receiving questions or requests entered by users through devices such as smartphones or computers from those devices.

[1387] The "means for analyzing the content of the received inquiry" is a function for understanding the content of the inquiry received from the user and performing processing to extract appropriate information.

[1388] The "means for recognizing the user's emotions based on the analyzed content" is a function for identifying the emotions the user is feeling based on the analysis results of the inquiry content.

[1389] The "means for generating an appropriate response according to the user's emotions" is a function that takes into account the recognized emotions of the user and creates an answer with a tone and content that is appropriate to those emotions.

[1390] The "means for providing the generated answer to the user" is a function for displaying or replying to the generated answer on the user's device.

[1391] The "means for receiving feedback on the provided answer" is a function that allows a user to input and receive an evaluation or additional comments on the provided answer.

[1392] "Means of using received feedback as learning data" is a function that accumulates feedback from users and uses it to improve the accuracy of future inquiries and to improve the system.

[1393] The "means for recommending movies and TV dramas" is a function for recommending appropriate movies and TV dramas to users based on the user's emotions and inquiries.

[1394] MODE FOR CARRYING OUT THE INVENTION

[1395] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates responses to the inquiries, and further recognizes the emotions of the users to provide appropriate responses. Hereinafter, specific embodiments of the present invention will be described.

[1396] Overall system configuration and functions

[1397] The system consists of the following main components:

[1398] 1. User Device:

[1399] It is a device with an interface that allows users to make inquiries through a smartphone application.

[1400] Hardware: Smartphone (e.g. iPhone, Android device)

[1401] Software: User Interface Application

[1402] 2. Server:

[1403] It is a central computer system that receives and processes user queries, analyzes the query, generates answers, provides feedback, and recognizes emotions.

[1404] Hardware: Server computer

[1405] Software: Web server, database server

[1406] 3. Natural Language Processing Engine:

[1407] This software analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to the next process.

[1408] Usage examples: SpaCy, NLTK

[1409] 4. Generative AI Models:

[1410] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[1411] Usage example: GPT-4, BERT

[1412] 5. Emotion Engine:

[1413] This software recognizes user emotions based on the content of the query and user input, and uses the recognized emotions to generate responses and adjust tone.

[1414] Usage examples: IBM Watson Tone Analyzer, Microsoft Azure Text Analytics

[1415] 6. Database:

[1416] It is a storage system for storing data on inquiries, generated answers, user feedback, and recognized emotions.

[1417] Usage example: MySQL, PostgreSQL

[1418] 7. Movie and TV series recommendation module:

[1419] This software recommends appropriate movies and dramas based on the user's emotions and inquiries.

[1420] Program processing

[1421] This program operates as follows:

[1422] 1. Receiving Inquiries

[1423] The user terminal receives an inquiry from the user through a smartphone app, which is then sent to the server.

[1424] 2. Analysis of inquiry content

[1425] The server sends the received inquiry to a natural language processing engine, which analyzes the inquiry and extracts keywords.

[1426] 3. User Emotion Recognition

[1427] The server receives the analysis results from the natural language processing engine and sends them to the emotion engine, which recognizes the user's emotions and returns the results to the server.

[1428] 4. Answer Generation

[1429] The server sends the analysis results and emotion recognition results to a generative AI model, which generates an answer appropriate to the user's emotions. For example, if the system recognizes that the user is "depressed," it will "recommend movies and TV shows that will help alleviate the depression."

[1430] Prompt Sentence Examples

[1431] If the user is in an emotionally depressed state, recommend movies or TV shows based on the following: "I'm feeling down. Can you recommend some movies that will cheer me up?"

[1432] 5. Providing generated answers

[1433] The server formats the generated answer, converts it into data for display on the user's device, and transmits it to the user. The user can view the answer through the user interface on their smartphone.

[1434] 6. Receive feedback and learn

[1435] The user terminal receives feedback from the user regarding the provided answer and sends it to the server, which stores this feedback in a database and uses it to improve future responses to inquiries.

[1436] In this way, the present invention can recognize the user's emotions and generate appropriate responses, thereby making it possible to improve the efficiency of inquiry response work and user satisfaction.

[1437] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1438] Step 1:

[1439] The user uses a smartphone app to input the inquiry and press the send button. An example input is "I'm feeling down, can you recommend a movie that will cheer me up?" The device then sends this inquiry to the server. The input data is the inquiry in text format, and the output is the data transferred to the server.

[1440] Step 2:

[1441] The server sends the received query content to a natural language processing engine, which analyzes the content and extracts key keywords (e.g., "feeling depressed," "cheerful," "movies"). The input data is the user's query content, and the output data is the analysis results including the extracted keywords.

[1442] Step 3:

[1443] The server sends the analysis results received from the natural language processing engine to the emotion engine. The emotion engine recognizes the user's emotions based on these analysis results. For example, it identifies emotions such as "feeling depressed." The input data is the analysis results, and the output data is the recognized emotional information.

[1444] Step 4:

[1445] The server sends the analysis results and emotion recognition results to the generative AI model. The generative AI model generates an answer that is appropriate to the user's emotions based on these inputs. For example, if the emotion is recognized as "depressed," it generates a recommendation such as "comedy movies" as an "uplifting movie." The input data are the analysis results and emotion recognition results, and the output data is the generated answer. The specific operation uses the following prompt sentence:

[1446] If the user is in an emotionally depressed state, recommend movies or TV shows based on the following: "I'm feeling down. Can you recommend some movies that will cheer me up?"

[1447] Step 5:

[1448] The server formats the generated answer and converts it into an appropriate data format (e.g., JSON, XML, HTML) for transmission to the device. The input data is the generated answer, and the output data is the formatted answer.

[1449] Step 6:

[1450] The terminal receives the formatted answer data sent from the server and displays it on the user interface. The user can check recommendations such as "Title A," which is a comedy movie, on the screen. The input data is the formatted answer data, and the output data is the displayed answer.

[1451] Step 7:

[1452] The user inputs and transmits feedback on the provided answer through the terminal. For example, the feedback may include "The recommendation for this movie was accurate." The terminal transmits this feedback to the server. The input data is the user's feedback, and the output data is the feedback transmitted to the server.

[1453] Step 8:

[1454] The server stores the feedback received from the device in a database and associates it with the analysis results and emotion recognition results. This feedback is used to improve future inquiries. The input data is the user feedback, and the output data is the stored feedback information.

[1455] 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.

[1456] 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.

[1457] 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.

[1458] [Fourth embodiment]

[1459] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1460] 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.

[1461] 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).

[1462] 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.

[1463] 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.

[1464] 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).

[1465] 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.

[1466] 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.

[1467] 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.

[1468] 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.

[1469] 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.

[1470] 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.

[1471] 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."

[1472] The present invention relates to an automatic response system that accepts inquiries from users, analyzes them, and generates and provides appropriate answers, thereby improving the efficiency of government office operations and improving public services. The following describes specific embodiments of the present invention.

[1473] Overall system configuration

[1474] The system consists of the following main components:

[1475] 1. User Device

[1476] It is a device that has an interface that allows citizens to make inquiries using government websites or applications.

[1477] 2. Server

[1478] A central computer system that receives and processes user inquiries, including analyzing the inquiry, generating answers, and receiving feedback.

[1479] 3. Natural Language Processing Engine

[1480] This is software or a program that analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to subsequent processing.

[1481] 4. Generative AI Models

[1482] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[1483] 5. Database

[1484] It is a storage system for saving data such as inquiry details, generated answers, and feedback from users.

[1485] Program processing

[1486] The program processing of this system will be explained in natural language below.

[1487] 1. Receiving Inquiries

[1488] Terminal: The user enters an inquiry about how to issue a resident registration card, etc., through the government website or application and presses the send button.

[1489] Server: Receives the inquiry sent from the terminal and temporarily stores it in a database.

[1490] 2. Analysis of inquiry content

[1491] Server: Sends the received query content to the natural language processing engine.

[1492] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident registration card," "issuance," and "method," and clarifies the subject's intent.

[1493] Server: Passes the analysis results to the generative AI model.

[1494] 3. Answer Generation

[1495] Generative AI model: Generates appropriate answers to user inquiries based on the analysis results of the natural language processing engine.

[1496] Generative AI model: Creates a response that includes detailed information and escalation as needed. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID."

[1497] 4. Providing generated answers

[1498] Server: Formats the answers received from the generative AI model and sends them to the device.

[1499] Terminal: Displays the generated answer to the user.

[1500] 5. Receiving Feedback

[1501] Terminal: The user enters feedback on the provided answer and presses the submit button.

[1502] Server: Receives feedback sent from devices and stores it in a database.

[1503] 6. Use of Feedback

[1504] Server: Passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy.

[1505] Specific examples

[1506] For example, when a user types "Please tell me how to issue a resident registration card" into a device and sends it, the server receives this and requests analysis from the natural language processing engine. Keywords such as "resident registration card," "issuance," and "method" are extracted as a result of the analysis. Based on this, the generative AI model generates a response such as "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and sends this to the user's device via the server.

[1507] In this way, the automatic response system based on the present invention can improve the efficiency of inquiries handled by government offices and the quality of services to citizens.

[1508] The processing flow will be explained below.

[1509] Step 1:

[1510] Terminal: The user enters an inquiry on the government website or application and presses the send button. For example, they might enter, "Please tell me how to issue a resident registration card."

[1511] Step 2:

[1512] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[1513] Step 3:

[1514] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[1515] Step 4:

[1516] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[1517] Step 5:

[1518] Server: Sends the analysis results received from the natural language processing engine to the generative AI model. Passes the analysis results to the generative AI model's API as input data.

[1519] Step 6:

[1520] Generative AI model: Generates an appropriate response based on the analysis results. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[1521] Step 7:

[1522] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[1523] Step 8:

[1524] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[1525] Step 9:

[1526] Terminal: The user enters feedback on the provided answer and presses the submit button. Feedback such as "The answer was easy to understand" or "I would like more specific information" is sent.

[1527] Step 10:

[1528] Server: Receives feedback sent from the device and stores it in a database. Passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy.

[1529] Example 1

[1530] 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."

[1531] Conventional government office inquiry response systems have struggled to provide fast and accurate responses to large volumes of inquiries, resulting in labor shortages and increased workloads. They also lacked a system for efficiently utilizing feedback on the answers provided to improve the accuracy of the system. Furthermore, while there is a need to respond to user inquiries in real time, there are also technical challenges in achieving this.

[1532] 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.

[1533] In this invention, the server includes means for receiving inquiries from users, means for saving the received inquiry content in a database, means for analyzing the inquiry content using a natural language processing engine, means for generating an answer using a generative AI model based on the analysis result, means for providing the generated answer to the user terminal, means for receiving user feedback and saving it in a database, and means for using the received feedback as learning data for the generative AI model to improve the accuracy of the answer, thereby enabling improved efficiency and accuracy in responding to inquiries and real-time user responses.

[1534] "Users" refers to citizens and residents who use the government's inquiry system.

[1535] "Inquiries" refer to questions or requests that users enter and submit through government websites or applications.

[1536] "Terminal" refers to a device used by a user to enter and send a query.

[1537] "Server" refers to a central computer system that receives and processes user queries.

[1538] A "natural language processing engine" refers to software or a program for analyzing the content of user inquiries.

[1539] A "generative AI model" refers to an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[1540] "Answer" refers to the appropriate response generated by the generative AI model based on the user's inquiry.

[1541] "Feedback" refers to the evaluation or opinion that a user enters in response to a provided answer.

[1542] A "database" refers to a storage system for storing data such as inquiries, generated responses, and user feedback.

[1543] "Training data" refers to data used to improve the performance of a generative AI model based on received feedback.

[1544] The present invention relates to an automatic response system that accepts inquiries from users, analyzes them, and generates and provides appropriate answers, thereby improving the efficiency of government office operations and improving public services. The following describes specific embodiments of the present invention.

[1545] Overall system configuration

[1546] The system consists of the following main components:

[1547] 1. User Device

[1548] This is a device with an interface that allows citizens to make inquiries using government websites or applications. This device can be any common computing device such as a personal computer (PC), smartphone, or tablet.

[1549] 2. Server

[1550] A central computer system that receives and processes user inquiries, analyzes the content of the inquiry, generates answers, and receives feedback. This server can be a cloud-based server or an on-premise server.

[1551] 3. Natural Language Processing Engine

[1552] This is software and a program for analyzing user inquiries. This engine analyzes the main keywords and intent of the inquiry and passes it on to subsequent processing. Specific software used is a Python natural language processing library (e.g., NLTK or spaCy).

[1553] 4. Generative AI Models

[1554] This is an artificial intelligence model that generates appropriate answers based on the analysis results of a natural language processing engine. This AI model is built using a machine learning framework (e.g., TensorFlow or PyTorch). The generative AI model continuously learns based on user feedback to improve the accuracy of its answers.

[1555] 5. Database

[1556] A storage system for saving data such as inquiry contents, generated answers, and user feedback, etc. It uses a relational database (e.g., MySQL, PostgreSQL) or a NoSQL database (e.g., MongoDB, DynamoDB).

[1557] Program processing

[1558] The program processing of this system is carried out in the following manner.

[1559] 1. Receiving Inquiries

[1560] Terminal: The user enters a question, such as how to issue a resident registration card, through a website or application and presses the send button. For example, the user enters a question such as "How do I issue a resident registration card?"

[1561] Server: Receives the inquiry sent from the user terminal and temporarily stores it in a database.

[1562] 2. Analysis of inquiry content

[1563] Server: Sends the received query content to the natural language processing engine.

[1564] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident registration card," "issuance," and "method," and returns the analysis results to the server.

[1565] 3. Answer Generation

[1566] Server: Passes the analysis results from the natural language processing engine to the generative AI model.

[1567] Generative AI model: Based on the analysis results, it generates a specific answer such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is identification."

[1568] 4. Providing generated answers

[1569] Server: Sends the answer received from the generative AI model to the user's device.

[1570] Terminal: Displays the generated answer to the user, for example, as text on a web page or in the form of a chatbot within an application.

[1571] 5. Receiving Feedback

[1572] Terminal: The user enters feedback on the provided answer and presses the submit button.

[1573] Server: Receives feedback sent from devices and stores it in a database.

[1574] 6. Use of Feedback

[1575] Server: The received feedback is used as learning data for the generative AI model. Based on this feedback data, the generative AI model re-learns and aims to improve the accuracy of its answers.

[1576] Specific examples

[1577] For example, suppose a user types and sends a query on their device, "Please tell me how to issue a resident registration card." The server receives this query and requests analysis from a natural language processing engine. Keywords such as "resident registration card," "issuance," and "method" are extracted as a result of the analysis. Based on this, the generative AI model generates a response: "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This response is sent to the user's device via the server. After seeing this response, the user enters feedback, which is received by the server and stored in a database. The generative AI model then re-learns based on the feedback, allowing it to provide more accurate responses to future inquiries.

[1578] Prompt Sentence Examples

[1579] "A user has submitted the inquiry '[How do I issue a residence certificate?]'. Please generate an appropriate answer based on this information."

[1580] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1581] Step 1:

[1582] Receiving inquiries

[1583] Terminal: The user enters the details of their inquiry (e.g., "Please tell me how to issue a resident registration card") through the government's website or application and presses the send button.

[1584] Input: Text data of the inquiry.

[1585] Server: Receives inquiries sent from user terminals and temporarily stores them in a database.

[1586] Output: The query is saved in the database.

[1587] Step 2:

[1588] Analysis of inquiry content

[1589] Server: Sends the received query content to the natural language processing engine.

[1590] Input: Text data of the inquiry content stored in the database.

[1591] Natural language processing engine: Analyzes the content of the inquiry and extracts keywords such as "resident certificate," "issuance," and "method." The analysis results are returned to the server.

[1592] Output: Analysis results including extracted keywords and query intent.

[1593] Step 3:

[1594] Generate answers

[1595] Server: Passes the analysis results from the natural language processing engine to the generative AI model.

[1596] Input: Analysis result data.

[1597] Generative AI model: Based on the analysis results, it generates appropriate answers to user inquiries. It generates specific answers such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID."

[1598] Output: The text data of the generated answers.

[1599] Step 4:

[1600] Providing generated answers

[1601] Server: Formats the answers received from the generative AI model and sends them to the user's device. To make the answers easier to view, the server formats the data in HTML or JSON format.

[1602] Input: Text data of the answer from the generative AI model.

[1603] Terminal: The generated answer is displayed to the user. For example, the answer can be displayed on a web page or in the form of a chatbot within an application.

[1604] Output: The answer displayed to the user.

[1605] Step 5:

[1606] Receiving Feedback

[1607] Terminal: The user enters feedback (e.g., "helpful" or "not helpful") for the provided answer and presses the submit button. It is also possible to enter specific comments.

[1608] Input: User feedback data.

[1609] Server: Receives feedback sent from devices and stores it in a database.

[1610] Output: Feedback data stored in a database.

[1611] Step 6:

[1612] Using Feedback

[1613] Server: Passes received feedback to the generative AI model.

[1614] Input: Feedback data stored in the database.

[1615] Generative AI model: The model is retrained based on the received feedback data to improve the accuracy of answers and the quality of responses.

[1616] Output: The retrained generative AI model.

[1617] The above is the specific program processing flow of the system, which improves the efficiency and accuracy of inquiries and enables real-time user support.

[1618] (Application example 1)

[1619] 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."

[1620] In modern electronic payment services, it is important to respond quickly and accurately to user inquiries. However, conventional systems take a long time to analyze the inquiry content and generate answers, and the answers provided are often not accurate enough. This makes it difficult to improve user satisfaction. Another problem is that feedback is not fully utilized, hindering system improvement. To solve these issues, there is a need for the development of a system that provides quick and accurate answers and efficiently utilizes feedback as training data.

[1621] 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.

[1622] In this invention, the server includes means for receiving inquiries from users, means for analyzing the received inquiries, means for generating appropriate answers based on the inquiries, means for providing the generated answers to the users, means for receiving feedback on the provided answers, means for using the received feedback as training data, means for identifying inquiries related to payments, and means for generating payment-related information. This enables quick and accurate analysis of inquiries and generation of answers in electronic payment services, thereby improving user satisfaction. Furthermore, by efficiently utilizing feedback as training data, the system can be continuously improved.

[1623] A "user inquiry" is the act of using an electronic device to ask a question or request information or a service.

[1624] "Means for receiving" refers to the functions and methods for obtaining information via electronic devices or communication systems.

[1625] "Means for analyzing the content of inquiries" refers to techniques and methods for understanding the meaning and intent of the information obtained and extracting relevant information.

[1626] "Means for generating appropriate answers" refers to techniques and methods for providing accurate and useful information to users' questions based on the analysis results.

[1627] "Means for providing the generated answer to the user" refers to the method or technology for displaying the generated answer on the user's terminal.

[1628] "Means for receiving feedback" refers to techniques and methods for receiving evaluations and opinions from users.

[1629] "Means of using feedback as training data" refers to techniques and methods for improving system performance based on received evaluations and opinions.

[1630] "Payment Inquiry" means a question or request from a user regarding the payment procedure or process.

[1631] "Means for generating payment information" refers to the technology or method for organizing and providing information related to payment procedures.

[1632] The present invention relates to an automatic response system that accepts and analyzes inquiries from users, and generates and provides appropriate responses, thereby improving the efficiency of electronic payment services and increasing user satisfaction. The following describes specific embodiments of the present invention.

[1633] Overall system configuration

[1634] The system consists of the following main components:

[1635] 1. User Device

[1636] A device with an interface that allows users to make inquiries using an electronic payment service application. Typical examples are smartphones and tablets.

[1637] 2. Server

[1638] A central computer system that receives inquiries, analyzes them, generates answers, and receives feedback. Often runs on a cloud-based server.

[1639] 3. Natural Language Processing Engine

[1640] This is software that analyzes the content of user inquiries and extracts key keywords and intent. Here, we use SpaCy (ja_core_news_sm model).

[1641] 4. Generative AI Models

[1642] This is an AI model that generates appropriate answers based on the analysis results of a natural language processing engine. It uses Huggingface's Transformers library.

[1643] 5. Database

[1644] It is a storage system for saving data such as inquiries, generated answers, and feedback from users.

[1645] System Program Processing

[1646] The server first receives the inquiry sent from the user's device. Next, it sends that content to a natural language processing engine (SpaCy) for analysis. The keywords and intent extracted as a result of the analysis are passed to a generative AI model (Transformers), which generates an appropriate answer. The generated answer is sent back to the user's device and provided to the user. When the user sends feedback on the answer provided, the server receives that feedback and stores it in a database. The received feedback is used as learning data for the generative AI model.

[1647] Specific examples

[1648] For example, if a user types "What should I do if my credit card payment doesn't work?" into a smartphone app and submits it, this inquiry is sent to the server. On the server, SpaCy's natural language processing engine extracts keywords such as "credit card," "payment," "doesn't work," and "how to fix it." Based on the analysis results, the generative AI model generates an answer such as "If your credit card payment doesn't work, please double-check your card information and try again. If the problem persists, please contact customer support," and provides it to the user.

[1649] Prompt Sentence Examples

[1650] What should I do if my credit card payment doesn't work?

[1651] The above is a specific embodiment for carrying out the present invention. This system enables quick and accurate analysis of user inquiries in electronic payment services and provides appropriate answers. Furthermore, by efficiently utilizing feedback as learning data, the system can be continuously improved.

[1652] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1653] Step 1:

[1654] User enters a query

[1655] A user uses a smartphone app for an electronic payment service to enter an inquiry and press the send button. As input, the user enters the question, "What should I do if my credit card payment doesn't go through?". As output, this inquiry is sent to the server.

[1656] Step 2:

[1657] The server receives the query

[1658] The server receives the inquiry sent from the user terminal. The input is text data (question) from the user. The output is to temporarily store the received inquiry in a database.

[1659] Step 3:

[1660] The server sends the query to the natural language processing engine

[1661] The server sends the received query to a natural language processing engine (SpaCy) for analysis. The query stored on the server is used as input. The analysis results (extracted keywords and intent) are obtained as output.

[1662] Step 4:

[1663] The natural language processing engine performs the analysis.

[1664] A natural language processing engine analyzes the content of an inquiry and extracts key keywords and intent. The input is text data (the question). The output is keywords such as "credit card," "payment," "not working," and "solution."

[1665] Step 5:

[1666] The server sends the analysis results to the generated AI model

[1667] The server sends the analysis results of the natural language processing engine to the generative AI model (Transformers). The input is the analysis results (extracted keywords and intent). The output is the generated answer.

[1668] Step 6:

[1669] Generative AI models generate answers

[1670] The generative AI model generates an appropriate answer based on the analysis results. The input is the analysis result, such as "My credit card payment is not working" and "What should I do?" The output is the answer, "If your credit card payment is not working, please double-check your card information and try again. If the problem persists, please contact customer support."

[1671] Step 7:

[1672] The server provides the generated answer to the user

[1673] The server formats the answer received from the generative AI model and sends it to the user's device. The input is the generated answer. The output is the answer displayed on the user's device.

[1674] Step 8:

[1675] Users can provide feedback on the answers provided

[1676] The user inputs feedback for the provided answer and presses the submit button. The input is the feedback content. The output is the feedback sent to the server.

[1677] Step 9:

[1678] Server receives feedback

[1679] The server receives the feedback sent from the user terminal and stores it in a database. As input, there is feedback data from the user. As output, the received feedback is stored in the database.

[1680] Step 10:

[1681] The server sends the feedback as training data to the generative AI model

[1682] The server passes the received feedback to the generative AI model and uses it to improve future inquiry responses and response accuracy. The input is the received feedback data. The output is an attempt to improve the performance of the generative AI model.

[1683] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1684] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates answers to the inquiries, and recognizes the emotions of the users to provide appropriate responses. Hereinafter, embodiments of the present invention will be described in detail.

[1685] Overall system configuration and functions

[1686] The system consists of the following main components:

[1687] 1. User Device

[1688] It is a device with an interface that allows users to make inquiries using government websites and applications.

[1689] 2. Server

[1690] It is a central computer system that receives and processes user queries, analyzes the query, generates answers, provides feedback, and recognizes emotions.

[1691] 3. Natural Language Processing Engine

[1692] This is software or a program that analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to the next process.

[1693] 4. Generative AI Models

[1694] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[1695] 5. Emotion Engine

[1696] It is software or a program that recognizes a user's emotions based on the content of the query or user input, and uses the recognized emotions to generate a response or adjust the tone.

[1697] 6. Database

[1698] It is a storage system for storing data on inquiries, generated answers, user feedback, and recognized emotions.

[1699] Program processing

[1700] The program processing of this system will be explained in natural language below.

[1701] 1. Receiving Inquiries

[1702] Terminal: The user enters an inquiry about how to issue a resident registration card, etc., through the government website or application and presses the send button.

[1703] 2. Analysis of inquiry content

[1704] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[1705] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[1706] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[1707] 3. User Emotion Recognition

[1708] Server: Sends the analysis results received from the natural language processing engine to the emotion engine. Calls the emotion engine's API and sends a request to recognize the user's emotion.

[1709] Emotion engine: Based on the user's input, it recognizes emotions such as excitement, anger, confusion, and joy and returns them to the server.

[1710] 4. Answer Generation

[1711] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model.

[1712] Generative AI model: Based on the emotion recognition results, it generates a response with a tone and content that matches the user's emotions. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[1713] 5. Providing generated answers

[1714] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[1715] 6. Displaying the Answer to the User

[1716] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[1717] 7. Receiving Feedback

[1718] Terminal: The user enters feedback on the provided answer and presses the submit button. Feedback such as "The answer was easy to understand" or "I would like more specific information" is sent.

[1719] 8. Use of Feedback and Emotional Data

[1720] Server: Receives feedback sent from the device and stores it in a database. The received feedback and emotional data are passed to the generative AI model, which uses it to improve future response and response accuracy.

[1721] Specific examples

[1722] For example, a user types, "Please tell me how to issue a resident registration card," and this is sent to the server. The server analyzes the content using a natural language processing engine and extracts keywords. At the same time, the emotion engine recognizes the user's emotion as "confusion." Based on the analysis results and emotion recognition results, the generative AI model generates a response in a tone that alleviates the "confusion," such as, "We will answer your questions about how to issue a resident registration card. Please come to the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This allows the user to resolve the problem with peace of mind.

[1723] In this way, the present invention can recognize the user's emotions and generate appropriate responses, thereby making it possible to improve the efficiency of government office inquiries and further enhance the quality of public services.

[1724] The processing flow will be explained below.

[1725] Step 1:

[1726] Terminal: The user uses the government's website or application to enter an inquiry such as "Please tell me how to issue a resident registration card" and presses the send button.

[1727] Step 2:

[1728] Server: Receives the inquiry sent by the user and temporarily stores it in a database. A unique identification number is assigned to the saved inquiry.

[1729] Step 3:

[1730] Server: Sends the received inquiry to the natural language processing engine. Calls the natural language processing engine's API and sends a request to analyze the inquiry.

[1731] Step 4:

[1732] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[1733] Step 5:

[1734] Server: Sends the analysis results received from the natural language processing engine to the emotion engine. Calls the emotion engine's API and sends a request to recognize the user's emotion.

[1735] Step 6:

[1736] Emotion engine: Based on the content of the inquiry, it recognizes the user's emotions such as "confusion," "anger," and "joy," and returns the emotional data to the server.

[1737] Step 7:

[1738] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model. Calls the generative AI model's API and sends a request to generate an appropriate answer.

[1739] Step 8:

[1740] Generative AI model: Based on the analysis results and emotion recognition results, it generates an appropriate answer to the question, "Please tell me how to issue a resident registration card." For example, it generates an answer such as, "To issue a resident registration card, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[1741] Step 9:

[1742] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[1743] Step 10:

[1744] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[1745] Step 11:

[1746] Device: The user enters their evaluation and opinion on the provided answer and presses the send button, sending a feedback message such as "The answer was easy to understand" or "I'd like to know more."

[1747] Step 12:

[1748] Server: Receives feedback sent from the device and stores it in a database. Passes the received feedback to the generative AI model and emotion engine, and uses it to improve future inquiry responses and response accuracy.

[1749] Example 2

[1750] 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."

[1751] Conventional automated response systems had the ability to provide appropriate answers to user inquiries, but they were unable to respond in a way that took the user's emotions into consideration. This made improving user satisfaction a challenge. Since the system lacked appropriate responses, particularly for users who were confused or angry, there was a need to further improve the quality of response to inquiries.

[1752] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1753] In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the content of the received inquiry, means for using a natural language processing engine to analyze the content of the inquiry, means for recognizing the user's emotions, means for generating an appropriate answer based on the analysis result and the emotion recognition result, means for providing the generated answer to the user, means for receiving feedback on the provided answer, and means for using the received feedback and emotion data as learning data. This enables appropriate responses and answers that take the user's emotions into consideration, improving the quality of inquiry responses and increasing user satisfaction.

[1754] "Means for receiving inquiries from users" refers to a function that allows the server to receive the content of inquiries entered by users through the government's website or application.

[1755] The "means for analyzing the content of the received inquiry" is a function that analyzes the content of the user's inquiry received by the server in order to understand it.

[1756] "Means for using a natural language processing engine" refers to a function that uses a software engine that performs natural language processing when analyzing the content of a received inquiry.

[1757] The "means for recognizing the user's emotions" is a function for identifying the emotional state at the time based on the content of the inquiry sent by the user.

[1758] "Means for generating appropriate answers based on analysis results and emotion recognition results" refers to a function for generating appropriate answers based on the results of natural language processing and emotion recognition.

[1759] "Means for providing the generated answer to the user" refers to the function by which the server sends the answer created by the generative AI model to the user's terminal and displays it.

[1760] The "means for receiving feedback on the provided answer" is a function that allows a user to input opinions and evaluations on the provided answer, and the server receives them.

[1761] The "means for using received feedback and emotion data as learning data" is a function for using the feedback and emotion recognition results received from the user to improve the system's performance and response accuracy.

[1762] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates answers to the inquiries, and recognizes the emotions of the users to provide appropriate responses. Hereinafter, embodiments of the present invention will be described in detail.

[1763] Overall system configuration and functions

[1764] The system consists of the following main components:

[1765] 1. User terminal: A device with an interface that allows users to make inquiries using government websites and applications. This includes PCs, smartphones, tablets, etc.

[1766] 2. Server: A central computer system that receives user queries, analyzes them, and generates answers. The server uses the following software and programs:

[1767] Natural language processing engines (e.g., SpaCy and BERT)

[1768] Emotion engine (e.g., IBM Watson Tone Analyzer)

[1769] Generative AI models (e.g., GPT-3)

[1770] Database (e.g. MySQL)

[1771] 3. Natural language processing engine: Software or a program that analyzes user inquiries, extracts the intent and keywords of the inquiry, and passes them on to the next process.

[1772] 4. Generative AI model: An artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of the natural language processing engine.

[1773] 5. Emotion Engine: Software or a program that recognizes the user's emotions based on the query or user input. The recognized emotions are used to generate responses and adjust the tone.

[1774] 6. Database: A storage system for storing data on queries, generated answers, user feedback, and recognized emotions.

[1775] Program processing flow

[1776] 1. Receiving Inquiries

[1777] Device: The user opens the government website or application, enters "Please tell me how to issue a resident registration card" into the inquiry form, and presses the send button.

[1778] 2. Analysis of inquiry content

[1779] Server: Receives the inquiry sent by the user and temporarily stores it in a local database. The stored inquiry is assigned a unique identification number (e.g., UUID).

[1780] Server: Sends the received query content to a natural language processing engine (e.g., SpaCy or BERT) by sending an HTTP POST request to the API endpoint of the natural language processing engine.

[1781] Natural language processing engine: Analyzes the content of the inquiry, extracts keywords such as "resident certificate," "issuance," and "method," and analyzes the sentence structure. The analysis results are returned to the server in JSON format.

[1782] 3. User Emotion Recognition

[1783] Server: Sends the analysis results received from the natural language processing engine to the emotion engine (e.g., IBM Watson Tone Analyzer). Sends an HTTP POST request to the emotion engine's API endpoint.

[1784] Emotion engine: Based on the user's input, it recognizes emotions such as excitement, anger, confusion, and joy and returns them to the server.

[1785] 4. Answer Generation

[1786] Server: Sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model (e.g., GPT-3).

[1787] Generative AI model: Based on the emotion recognition results, it generates a response with a tone and content that matches the user's emotions. For example, it generates a response such as, "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card," and returns it to the server.

[1788] 5. Providing generated answers

[1789] Server: Formats the answers received from the generative AI model and converts them into data for sending to the user device, for example, into HTML or JSON format.

[1790] 6. Displaying the Answer to the User

[1791] Terminal: Receives the response data sent from the server and displays it on the user interface. The user confirms the response on the screen: "To issue a resident registration certificate, please visit the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card."

[1792] Specific examples

[1793] For example, a user types, "Please tell me how to issue a resident registration card," and this is sent to the server. The server uses a natural language processing engine to analyze the content and extract keywords. At the same time, the emotion engine recognizes the user's emotion as "confusion." Based on this analysis result and the emotion recognition result, the generative AI model generates a response in a tone that alleviates the "confusion," such as, "We will answer your questions about how to issue a resident registration card. Please come to the relevant counter between 9:00 and 17:00 on weekdays. The required document is an ID card." This allows the user to resolve the problem with peace of mind.

[1794] Prompt Sentence Examples

[1795] example:

[1796] User's question: "How do I get a certificate of residence issued?"

[1797] Analysis results: {"keywords": ["resident card", "issuance", "method"], "intent": "information_request"}

[1798] Emotion recognition result: "confused"

[1799] Based on this prompt, a generative AI model generates an appropriate response, and the system as a whole accumulates feedback and sentiment data that can be used to improve future inquiries.

[1800] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1801] Step 1:

[1802] The user opens the city hall's website or application, enters "Please tell me how to issue a resident registration card" into the inquiry form, and presses the send button.

[1803] Input: User's inquiry: "Please tell me how to issue a resident registration card."

[1804] Specific operation: When the user clicks the "Send" button, the terminal sends the input contents to the server as an HTTP POST request.

[1805] Output: The query received by the server.

[1806] Step 2:

[1807] The server receives the query sent by the user and temporarily stores it in a local database, where it is assigned a unique identification number (e.g., UUID).

[1808] Input: User's inquiry: "Please tell me how to issue a resident registration card."

[1809] Specific operation: The server saves the inquiry in the "inquiries" table of the MySQL database and issues a unique ID.

[1810] Output: The query and its unique identification number stored in the database.

[1811] Step 3:

[1812] The server sends the received query to a natural language processing engine.

[1813] Input: The query and its unique identification number stored in the database.

[1814] Specific operation: The server converts the query content into JSON format and sends an HTTP POST request to the API endpoint of a natural language processing engine (e.g., SpaCy or BERT).

[1815] Output: The results of the analysis by the natural language processing engine are received in JSON format.

[1816] Step 4:

[1817] The natural language processing engine analyzes the inquiry, extracts keywords such as "resident registration," "issuance," and "method," and analyzes the sentence structure.

[1818] Input: JSON data of the query sent from the server.

[1819] Specific operation: The natural language processing engine performs text analysis, compiles the extracted keywords and intent into JSON format, and returns it to the server.

[1820] Output: JSON format data containing the parsed results.

[1821] Step 5:

[1822] The server sends the analysis results from the natural language processing engine to the emotion engine.

[1823] Input: JSON data of the analysis results obtained from the natural language processing engine.

[1824] Specific operation: The server sends the analysis results to the API endpoint of the emotion engine (e.g., IBM Watson Tone Analyzer) via an HTTP POST request.

[1825] Output: Receives emotion data recognized by the emotion engine in JSON format.

[1826] Step 6:

[1827] The emotion engine recognizes the user's emotions based on the content of the query and identifies emotions such as "confusion."

[1828] Input: JSON data of the analysis result sent from the server.

[1829] Specific operation: The emotion engine performs emotion analysis and returns the recognized emotion tag to the server.

[1830] Output: JSON formatted data containing emotion recognition results.

[1831] Step 7:

[1832] The server sends the analysis results and emotion recognition results received from the natural language processing engine and emotion engine to the generative AI model.

[1833] Input: JSON data of analysis results and emotion recognition results obtained from the natural language processing engine and emotion engine.

[1834] Specific operation: The server sends a prompt message to the API endpoint of the generative AI model (e.g., GPT-3) based on the analysis results and emotion recognition results.

[1835] Output: JSON-formatted data containing the answers generated by the generative AI model.

[1836] Step 8:

[1837] The generative AI model generates responses with a tone and content that matches the user's emotions based on the emotion recognition results.

[1838] Input: Prompt sentence based on analysis results and emotion recognition results.

[1839] Specific operation: The generative AI model generates an appropriate answer and returns it to the server in JSON format.

[1840] Output: JSON-formatted data containing the appropriate answer from the generative AI model.

[1841] Step 9:

[1842] The server formats the answers received from the generative AI model and converts them into data for transmission to the user's device.

[1843] Input: JSON data of the appropriate answer data from the generative AI model.

[1844] What happens: The server embeds the response in an HTML template and reformats it for display in the user's browser.

[1845] Output: Formatted response data.

[1846] Step 10:

[1847] The user terminal receives the response data sent from the server and displays it on the user interface.

[1848] Input: Formatted response data.

[1849] Specific operation: The web browser on the user's device renders the received HTML data and displays it to the user.

[1850] Output: Answers displayed on the screen.

[1851] Step 11:

[1852] The user enters feedback on the provided answer and presses the submit button.

[1853] Input: The feedback entered by the user.

[1854] Specific operation: When a user fills out the feedback form and presses the submit button, the terminal sends the feedback to the server as an HTTP POST request.

[1855] Output: The feedback received by the server.

[1856] Step 12:

[1857] The server receives the feedback sent from the device and stores it in a database. The received feedback and emotional data are used as training data for the generative AI model.

[1858] Input: User feedback.

[1859] What it does: The server stores the feedback in a "feedback" table in the database and uses it for subsequent analysis and improvement processes.

[1860] Output: Feedback content and sentiment data stored in a database.

[1861] (Application example 2)

[1862] 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."

[1863] Conventional query processing systems do not recognize the user's emotions and generate answers based on them, but simply respond based on keywords, which can result in users not receiving satisfactory answers. Furthermore, when a user requests recommendations for content such as movies or TV dramas, there is a problem in that appropriate recommendations based on the user's emotions cannot be made.

[1864] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for receiving an inquiry from a user, means for analyzing the received inquiry content, means for recognizing the user's emotions based on the analyzed content, means for generating an appropriate answer according to the user's emotions, means for providing the generated answer to the user, means for receiving feedback on the provided answer, means for using the received feedback as learning data, and means for recommending movies and TV shows. This makes it possible to respond to inquiries and recommend appropriate movies and TV shows that take the user's emotions into consideration.

[1865] "Means for receiving inquiries from users" refers to a function for receiving questions or requests entered by users through devices such as smartphones or computers from those devices.

[1866] The "means for analyzing the content of the received inquiry" is a function for understanding the content of the inquiry received from the user and performing processing to extract appropriate information.

[1867] The "means for recognizing the user's emotions based on the analyzed content" is a function for identifying the emotions the user is feeling based on the analysis results of the inquiry content.

[1868] The "means for generating an appropriate response according to the user's emotions" is a function that takes into account the recognized emotions of the user and creates an answer with a tone and content that is appropriate to those emotions.

[1869] The "means for providing the generated answer to the user" is a function for displaying or replying to the generated answer on the user's device.

[1870] The "means for receiving feedback on the provided answer" is a function that allows a user to input and receive an evaluation or additional comments on the provided answer.

[1871] "Means of using received feedback as learning data" is a function that accumulates feedback from users and uses it to improve the accuracy of future inquiries and to improve the system.

[1872] The "means for recommending movies and TV dramas" is a function for recommending appropriate movies and TV dramas to users based on the user's emotions and inquiries.

[1873] MODE FOR CARRYING OUT THE INVENTION

[1874] The present invention relates to an automatic response system that receives inquiries from users, analyzes the inquiries, generates responses to the inquiries, and further recognizes the emotions of the users to provide appropriate responses. Hereinafter, specific embodiments of the present invention will be described.

[1875] Overall system configuration and functions

[1876] The system consists of the following main components:

[1877] 1. User Device:

[1878] It is a device with an interface that allows users to make inquiries through a smartphone application.

[1879] Hardware: Smartphone (e.g. iPhone, Android device)

[1880] Software: User Interface Application

[1881] 2. Server:

[1882] It is a central computer system that receives and processes user queries, analyzes the query, generates answers, provides feedback, and recognizes emotions.

[1883] Hardware: Server computer

[1884] Software: Web server, database server

[1885] 3. Natural Language Processing Engine:

[1886] This software analyzes the content of user inquiries, extracting the intent and keywords of the inquiry and passing them on to the next process.

[1887] Usage examples: SpaCy, NLTK

[1888] 4. Generative AI Models:

[1889] This is an artificial intelligence model that generates appropriate answers to user inquiries based on the analysis results of a natural language processing engine.

[1890] Usage example: GPT-4, BERT

[1891] 5. Emotion Engine:

[1892] This software recognizes user emotions based on the content of the query and user input, and uses the recognized emotions to generate responses and adjust tone.

[1893] Usage examples: IBM Watson Tone Analyzer, Microsoft Azure Text Analytics

[1894] 6. Database:

[1895] It is a storage system for storing data on inquiries, generated answers, user feedback, and recognized emotions.

[1896] Usage example: MySQL, PostgreSQL

[1897] 7. Movie and TV series recommendation module:

[1898] This software recommends appropriate movies and dramas based on the user's emotions and inquiries.

[1899] Program processing

[1900] This program operates as follows:

[1901] 1. Receiving Inquiries

[1902] The user terminal receives an inquiry from the user through a smartphone app, which is then sent to the server.

[1903] 2. Analysis of inquiry content

[1904] The server sends the received inquiry to a natural language processing engine, which analyzes the inquiry and extracts keywords.

[1905] 3. User Emotion Recognition

[1906] The server receives the analysis results from the natural language processing engine and sends them to the emotion engine, which recognizes the user's emotions and returns the results to the server.

[1907] 4. Answer Generation

[1908] The server sends the analysis results and emotion recognition results to a generative AI model, which generates an answer appropriate to the user's emotions. For example, if the system recognizes that the user is "depressed," it will "recommend movies and TV shows that will help alleviate the depression."

[1909] Prompt Sentence Examples

[1910] If the user is in an emotionally depressed state, recommend movies or TV shows based on the following: "I'm feeling down. Can you recommend some movies that will cheer me up?"

[1911] 5. Providing generated answers

[1912] The server formats the generated answer, converts it into data for display on the user's device, and transmits it to the user. The user can view the answer through the user interface on their smartphone.

[1913] 6. Receive feedback and learn

[1914] The user terminal receives feedback from the user regarding the provided answer and sends it to the server, which stores this feedback in a database and uses it to improve future responses to inquiries.

[1915] In this way, the present invention can recognize the user's emotions and generate appropriate responses, thereby making it possible to improve the efficiency of inquiry response work and user satisfaction.

[1916] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1917] Step 1:

[1918] The user uses a smartphone app to input the inquiry and press the send button. An example input is "I'm feeling down, can you recommend a movie that will cheer me up?" The device then sends this inquiry to the server. The input data is the inquiry in text format, and the output is the data transferred to the server.

[1919] Step 2:

[1920] The server sends the received query content to a natural language processing engine, which analyzes the content and extracts key keywords (e.g., "feeling depressed," "cheerful," "movies"). The input data is the user's query content, and the output data is the analysis results including the extracted keywords.

[1921] Step 3:

[1922] The server sends the analysis results received from the natural language processing engine to the emotion engine. The emotion engine recognizes the user's emotions based on these analysis results. For example, it identifies emotions such as "feeling depressed." The input data is the analysis results, and the output data is the recognized emotional information.

[1923] Step 4:

[1924] The server sends the analysis results and emotion recognition results to the generative AI model. The generative AI model generates an answer that is appropriate to the user's emotions based on these inputs. For example, if the emotion is recognized as "depressed," it generates a recommendation such as "comedy movies" as an "uplifting movie." The input data are the analysis results and emotion recognition results, and the output data is the generated answer. The specific operation uses the following prompt sentence:

[1925] If the user is in an emotionally depressed state, recommend movies or TV shows based on the following: "I'm feeling down. Can you recommend some movies that will cheer me up?"

[1926] Step 5:

[1927] The server formats the generated answer and converts it into an appropriate data format (e.g., JSON, XML, HTML) for transmission to the device. The input data is the generated answer, and the output data is the formatted answer.

[1928] Step 6:

[1929] The terminal receives the formatted answer data sent from the server and displays it on the user interface. The user can check recommendations such as "Title A," which is a comedy movie, on the screen. The input data is the formatted answer data, and the output data is the displayed answer.

[1930] Step 7:

[1931] The user inputs and transmits feedback on the provided answer through the terminal. For example, the feedback may include "The recommendation for this movie was accurate." The terminal transmits this feedback to the server. The input data is the user's feedback, and the output data is the feedback transmitted to the server.

[1932] Step 8:

[1933] The server stores the feedback received from the device in a database and associates it with the analysis results and emotion recognition results. This feedback is used to improve future inquiries. The input data is the user feedback, and the output data is the stored feedback information.

[1934] 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.

[1935] 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.

[1936] 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.

[1937] 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.

[1938] 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.

[1939] 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.

[1940] 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).

[1941] 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.

[1942] 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."

[1943] 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.

[1944] 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).

[1945] 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.

[1946] 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.

[1947] 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.

[1948] 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.

[1949] 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.

[1950] 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.

[1951] 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.

[1952] 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.

[1953] 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.

[1954] 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.

[1955] The following is further disclosed regarding the above embodiment.

[1956] (Claim 1)

[1957] means for receiving a query from a user;

[1958] means for analyzing the received inquiry;

[1959] means for generating an appropriate response based on the query;

[1960] means for providing the generated answer to the user;

[1961] a means for receiving feedback on the answers provided;

[1962] a means for utilizing the received feedback as training data; and

[1963] A system including:

[1964] (Claim 2)

[1965] 2. The system according to claim 1, wherein the content of the user's inquiry is analyzed using a natural language processing engine.

[1966] (Claim 3)

[1967] The system of claim 1, wherein an answer is generated using a generative AI model based on the analysis results.

[1968] "Example 1"

[1969] (Claim 1)

[1970] means for receiving a query from a user;

[1971] a means for storing the received inquiry content in a database;

[1972] A means for analyzing the content of the inquiry using a natural language processing engine;

[1973] A means for generating an answer using a generative AI model based on the analysis results;

[1974] means for providing the generated answer to a user terminal;

[1975] means for receiving and storing user feedback in a database;

[1976] A means to use the received feedback as training data for the generative AI model to improve the accuracy of the answers;

[1977] A system including:

[1978] (Claim 2)

[1979] 2. The system according to claim 1, wherein the system processes the inquiry content and analysis results sequentially and provides a response in real time.

[1980] (Claim 3)

[1981] The system of claim 1, wherein the generative AI model is provided with feedback data and retrained to improve the quality of the answers.

[1982] "Application Example 1"

[1983] (Claim 1)

[1984] means for receiving a query from a user;

[1985] means for analyzing the received inquiry;

[1986] means for generating an appropriate response based on the query;

[1987] means for providing the generated answer to the user;

[1988] a means for receiving feedback on the answers provided;

[1989] a means for utilizing the received feedback as training data; and

[1990] a means of identifying payment inquiries;

[1991] means for generating payment information;

[1992] A system including:

[1993] (Claim 2)

[1994] 2. The system according to claim 1, wherein the content of the user's inquiry is analyzed using a natural language processing engine.

[1995] (Claim 3)

[1996] The system of claim 1, wherein an answer is generated using a generative AI model based on the analysis results.

[1997] "Example 2: Combining Emotion Engines"

[1998] (Claim 1)

[1999] means for receiving a query from a user;

[2000] means for analyzing the received inquiry;

[2001] a means for using a natural language processing engine to analyze the query;

[2002] means for recognizing a user's emotion;

[2003] A means for generating an appropriate answer based on the analysis result and the emotion recognition result;

[2004] means for providing the generated answer to the user;

[2005] a means for receiving feedback on the answers provided;

[2006] a means for utilizing the received feedback and emotion data as training data;

[2007] A system including:

[2008] (Claim 2)

[2009] 2. The system according to claim 1, wherein when analyzing the content of a user's inquiry, the system uses a natural language processing engine to analyze the content and an emotion engine to recognize emotions.

[2010] (Claim 3)

[2011] The system of claim 1, wherein an answer is generated using a generative AI model based on the analysis results and emotion recognition results.

[2012] "Application example 2 when combining emotion engines"

[2013] (Claim 1)

[2014] means for receiving a query from a user;

[2015] means for analyzing the received inquiry;

[2016] means for recognizing a user's emotion based on the analyzed content;

[2017] A means for generating an appropriate response according to the user's feelings;

[2018] means for providing the generated answer to the user;

[2019] a means for receiving feedback on the answers provided;

[2020] a means for utilizing the received feedback as training data; and

[2021] A means of recommending movies and TV shows,

[2022] A system including:

[2023] (Claim 2)

[2024] 2. The system according to claim 1, wherein the content of the user's inquiry is analyzed using a natural language processing engine.

[2025] (Claim 3)

[2026] The system of claim 1, which generates an answer using a generative AI model based on the analysis results and the user's emotion recognition results. [Explanation of symbols]

[2027] 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. means for receiving a query from a user; means for analyzing the received inquiry; means for generating an appropriate response based on the query; means for providing the generated answer to the user; a means for receiving feedback on the answers provided; a means for utilizing the received feedback as training data; and A system including:

2. 2. The system according to claim 1, wherein the content of the user's inquiry is analyzed using a natural language processing engine.

3. The system of claim 1 , wherein an answer is generated using a generative AI model based on the analysis results.

Images