system

Abstract

We provide the system. [Solution] Means of collecting industry-specific data, A means for preprocessing the aforementioned data and training a machine learning model, A means for automatically generating frequently asked questions and their answers using the aforementioned trained model, A means for receiving questions from users and providing relevant answers by referring to the generated question and answer database, The system includes means for supplementing answers to questions not present in the aforementioned database using a generative AI model.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In industries that require expertise, customers and employees often pose complex and detailed questions, and answering them quickly and accurately requires a lot of resources. Currently, manual answer creation is the mainstream, which has problems of heavy human burden, lack of guarantee of answer consistency and immediacy. Furthermore, in industries where information is frequently updated, it is difficult to maintain answers based on the latest information.

Means for Solving the Problems

[0005] This invention provides a system for automatically generating and managing FAQs for specialized questions by collecting industry-specific data and training an AI model based on that data. Specifically, it includes means for collecting the latest information using online databases and APIs, and means for preprocessing the collected data and training a machine learning model to acquire specialized knowledge. Furthermore, it includes means for automatically generating frequently asked questions and their answers using this trained model, and for providing relevant answers to user questions by referring to the generated database. This makes it possible to provide accurate and rapid answers without human intervention, thereby improving operational efficiency and reducing the burden on personnel.

[0006] "Industry-specific data" refers to information and knowledge related to a particular industry, including industry processes, regulations, trends, and terminology.

[0007] "Preprocessing" is the process of converting raw data into a format suitable for analysis or input to machine learning models, and involves tasks such as noise reduction, data standardization, and deduplicate removal.

[0008] A "machine learning model" is a collection of algorithms that learn patterns from data and use them to make predictions and classifications.

[0009] A "generative AI model" is an artificial intelligence that has the ability to generate natural language from large datasets, and is a model trained to create sentences like a human.

[0010] "FAQ" refers to a collection of frequently asked questions and their answers.

[0011] A "database" is a digital storage system that systematically organizes information, making it easy to search and update.

[0012] An "online database" is a data storage service accessible via the internet, containing a collection of various types of information.

[0013] "API" stands for Application Programming Interface, and it is an interface that allows software to exchange data and functions with each other.

[0014] "Complementing an answer" is the process of generating or adding additional information to fill in incomplete information or missing data, in order to provide a complete answer. [Brief explanation of the drawing]

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

Mode for Carrying Out the Invention

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

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

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

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

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

[0021] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0022] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0023] [First Embodiment]

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

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

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

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

[0028] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0030] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0032] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

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

[0034] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

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

[0036] This invention is an AI-based FAQ system that provides accurate and rapid answers to frequently asked questions in a specific industry. The system collects industry-specific data and uses machine learning models to train its expertise. It also employs generative AI to generate answers to user questions and maintains and manages the database.

[0037] Data collection and utilization

[0038] The server periodically collects industry-specific information through online databases and APIs. For example, in the medical industry, it would retrieve the latest medical papers and guidelines; in the legal industry, it would retrieve case law and legal amendment information. This ensures that the data is always up-to-date.

[0039] Model training and FAQ generation

[0040] The server preprocesses the collected data and inputs it into a machine learning model. The model learns industry-specific patterns and trends from the collected information and uses this to generate an FAQ database. This automatically creates questions and answers that reflect industry expertise.

[0041] User Interface Support

[0042] Users input questions in natural language from their devices and send them to the server. The server searches its FAQ database for relevant answers and provides them. Even if there is no matching answer in the FAQ database, it uses a generative AI model to supplement it with an appropriate answer. For example, if a user inputs "Tell me about new diabetes treatments" into their device, they can instantly receive a detailed answer based on the latest information.

[0043] This system functions as a specialized question-answering system tailored to specific industries, enabling the rapid and accurate provision of information while minimizing the need for significant human intervention.

[0044] The following describes the processing flow.

[0045] Step 1:

[0046] The server periodically collects industry-specific data from online databases and APIs. This includes collecting the latest news articles, research literature, and regulatory information, and storing it in the database.

[0047] Step 2:

[0048] The server preprocesses the collected data. Specifically, it standardizes the data format, removes noise, and eliminates duplicate information. This ensures that subsequent processing proceeds smoothly.

[0049] Step 3:

[0050] The server uses pre-processed data to input into a machine learning model, allowing it to learn its expertise. This process adjusts the model to understand key patterns and the meaning of technical terms within the data.

[0051] Step 4:

[0052] The server automatically generates FAQs using a pre-trained model. It predicts common questions related to the industry and generates detailed answers to them.

[0053] Step 5:

[0054] The user enters a specific question from their device. For example, they might enter a question in natural language such as, "I want to know about the latest tax law revisions."

[0055] Step 6:

[0056] The terminal formats the user's question appropriately and sends it to the server. The server searches the FAQ database based on the question.

[0057] Step 7:

[0058] The server selects the most relevant answer from the FAQ database and provides it to the user. If no relevant FAQ exists, it uses a generative AI model to generate a new answer.

[0059] Step 8:

[0060] The terminal displays the response received from the server to the user. The user can review the response on the screen and enter further questions if necessary.

[0061] This series of steps allows users to quickly obtain the information they need.

[0062] (Example 1)

[0063] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0064] Providing quick and accurate answers to frequently asked questions within specific industries is becoming increasingly difficult due to the sheer volume of information and the growing demand for speed. Traditional systems struggle to meet user demands because information updates and the quality of responses are insufficient. Furthermore, the scattered nature of information sources necessitates a means of accurately collecting and efficiently processing relevant information.

[0065] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0066] In this invention, the server includes information aggregation means, means for processing the information and training it using a machine learning algorithm, and means for automatically generating frequently asked questions and their answers using the trained algorithm. This makes it possible to efficiently collect and analyze industry-specific information and provide users with quick and accurate answers.

[0067] "Information aggregation means" refers to methods for automatically acquiring information related to a specific industry through network databases or programmatic interfaces.

[0068] A "machine learning algorithm" is a collection of computer programs designed to learn specific patterns or knowledge from large amounts of data.

[0069] A "trained algorithm" is an algorithm that has been trained by a machine learning algorithm and possesses the ability to generate answers for a specific task.

[0070] A "generative AI algorithm" is an artificial intelligence technology that can generate new information and data based on natural language processing techniques.

[0071] "Information source" refers to a database or knowledge base where answers to user questions are accumulated in the form of FAQs or similar.

[0072] To implement this invention, the server primarily uses information aggregation means, machine learning algorithms, and generative AI algorithms. The server is equipped with information aggregation means for acquiring industry-specific information from online databases and APIs via a network. For example, in the medical industry, the server can periodically collect the latest medical research papers and guidelines.

[0073] The collected information is processed by a preprocessing module on the server. This preprocessing includes tasks such as data cleaning, text normalization, and tokenization. The resulting clean data is then input into a deep learning-based machine learning algorithm to learn industry-specific patterns and trends.

[0074] The trained model is used to quickly generate answers to user questions. Users input questions in natural language through their devices, and the server receives them. The server searches the FAQ database and selects an appropriate answer. If a suitable answer is not found in the database, the server utilizes a generative AI algorithm to generate a new answer. The generative AI uses natural language processing techniques to provide answers in context relevant to the user's question.

[0075] For example, even if a user sends a question from their device such as "Tell me about new diabetes treatments," the server can immediately provide a detailed answer based on the latest medical information. An example of a prompt message would be, "The user is asking 'Tell me about new diabetes treatments.' Please generate a detailed answer based on the latest medical data."

[0076] In this way, the server can provide users with quick and accurate information and propose useful solutions tailored to their needs.

[0077] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0078] Step 1:

[0079] The server uses information aggregation tools to retrieve industry-specific information from the network. Specifically, the server issues HTTP requests to online databases and APIs to download the latest research papers and guidelines. The input to this process is the URL of the information source or the API endpoint, and the output is stored as raw data.

[0080] Step 2:

[0081] The server preprocesses the acquired raw data. Specifically, the server runs a data cleaning algorithm to remove noise and inconsistencies. Text normalization and tokenization are also performed. The input to this process is the raw data obtained in step 1, and the output is clean data that can be processed by a machine learning algorithm.

[0082] Step 3:

[0083] The server trains a machine learning algorithm using clean data. The server inputs the data into the model, allowing it to learn patterns and trends in a specific domain. At this stage, the input is the clean data generated in step 2, and the output is the trained model.

[0084] Step 4:

[0085] The terminal receives a question from the user and sends that question to the server. Specifically, the user enters the question in natural language into the terminal and clicks the "Send" button. The input in this process is the user's question, and the output is sent as a request to the server.

[0086] Step 5:

[0087] The server first searches the FAQ database for the received question to find a relevant answer. If no relevant answer is found, the server uses a generative AI model to create a new answer. The input for this step is the question obtained in step 4, and the output is either an answer from the FAQ database or a new answer from the generative AI model.

[0088] Step 6:

[0089] The server provides the user with the generated or retrieved answer. Specifically, the server composes the answer and sends it to the terminal. The input to this process is the answer data from step 5, and the output is the answer displayed on the user's terminal.

[0090] (Application Example 1)

[0091] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0092] In physical stores, customers often want to quickly and accurately inquire about product information, sales status, and promotions, but directly asking store staff is often time-consuming. Furthermore, immediate responses are difficult when staff are unavailable. This poses a risk of negatively impacting the customer experience.

[0093] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0094] In this invention, the server includes means for collecting industry-specific information, means for preprocessing the information and training it using a machine learning algorithm, means for automatically generating frequently asked inquiries and their answers using the trained algorithm, and means for customers to query in-store information from a mobile terminal and display the responses on the mobile terminal. This allows customers to quickly obtain necessary information using a mobile device without going through store staff, thereby improving the customer experience.

[0095] "Industry-specific information" refers to information related to a particular industry, and is knowledge and data that is only useful within that industry.

[0096] "Preprocessing" refers to the initial data processing steps performed to prepare raw data into a format that can be easily handled by machine learning algorithms.

[0097] A "machine learning algorithm" refers to a technique that uses data to train a model and then makes predictions or classifications based on new data.

[0098] A "pre-trained algorithm" is a machine learning model that has already been trained and is ready to handle a specific task.

[0099] An "inquiry" refers to a question or request made by a user to obtain information.

[0100] The "Answer Information Repository" is a database that stores past inquiries and their corresponding answers.

[0101] A "generative AI algorithm" is an artificial intelligence technology that generates natural language responses based on input data.

[0102] "Mobile devices" refer to portable computer devices such as smartphones and tablets.

[0103] "In-store information" refers to business information related to a specific physical store, such as products, inventory, and promotions.

[0104] The system for implementing this invention consists of a server, a customer's mobile terminal, and a store information management system. The server collects industry-specific information through online databases and information provision interfaces and preprocesses it. This prepares the data in a format that can be efficiently handled by machine learning algorithms. The preprocessed data is then trained by the server using machine learning algorithms and used to automatically generate inquiries and their answers.

[0105] The trained algorithm uses natural language processing techniques to accurately understand industry-specific terminology and concepts and generate responses. When a customer makes a query using a mobile device, the server refers to the relevant query and answer database to provide a corresponding response. If a matching answer does not exist in the database, the generation AI algorithm generates a supplementary answer in real time and displays it on the customer's device.

[0106] For example, if a customer in a store enters "Please tell me about the current special sale in the store" into a mobile terminal, the server retrieves the latest sale information from its database and provides an accurate response based on that information.

[0107] An example of a prompt for the generative AI model in this case is as follows:

[0108] "Could you tell me about any sales currently running in your store? I'd like to know more about the latest discounts and campaigns."

[0109] This allows customers to quickly and accurately obtain the information they need without having to ask store staff directly.

[0110] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0111] Step 1:

[0112] The server collects industry-specific information through online databases and information provision interfaces. It retrieves information from specified APIs as input and converts unstructured data into an easily understandable format. The output is data that has been prepared for preprocessing.

[0113] Step 2:

[0114] The server preprocesses the collected data. The input is the data prepared in step 1. Data cleaning, tokenization, and feature extraction are performed to transform the data into a format usable by machine learning algorithms. Noise reduction and filtering of necessary data are also performed during this process. The output becomes the training dataset.

[0115] Step 3:

[0116] The server uses pre-processed data to train a machine learning algorithm. It uses a training dataset as input and learns patterns using algorithms such as neural networks. The output is a trained algorithm that can be used to generate queries and answers.

[0117] Step 4:

[0118] A user enters a specific query from a mobile device. The input is natural language text received through the device's interface. The device sends the entered text to the server. The output is the request to the server.

[0119] Step 5:

[0120] The server receives a query from a user and consults its answer database. The input is text sent by the user. The server uses natural language processing techniques to search the database for existing answers related to the query. The output is the answer as a search result.

[0121] Step 6:

[0122] If the answer does not exist in the answer database, the server uses a generative AI model to complete the answer. The input is the query for which no match was found in step 5. The generative AI model uses a prompt to generate a response in natural language. The output is the generated answer text.

[0123] Step 7:

[0124] The server returns the results to the user's terminal. The input is the answer obtained in step 5 or step 6. The terminal displays the received answer on the screen and informs the user. The output is the information displayed to the user.

[0125] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0126] This invention realizes a system that provides more personalized answers by combining an AI-based FAQ system using industry-specific data with a function that recognizes user emotions. This system comprises a server, terminals, and an emotion engine, and provides industry-savvy answers instantly while adjusting the tone and content according to the user's emotional state.

[0127] Data collection and model learning

[0128] The server collects industry-specific data through online databases and APIs. The collected data is preprocessed, and industry expertise is learned using machine learning models. These models utilize natural language processing techniques to understand technical terms and complex concepts.

[0129] FAQ generation and management

[0130] The server uses a pre-trained model to automatically generate frequently asked questions and their answers. The generated FAQs are stored in a database and regularly updated based on the latest information.

[0131] Emotion recognition and response

[0132] When a user enters a question via their device, the emotion engine analyzes the user's wording and, if possible, their tone of voice to identify their emotions. For example, if the user is feeling anxious, the engine adjusts the response to alleviate those emotions.

[0133] User response

[0134] The server selects the most relevant answer from the FAQ database, but adjusts the tone of the answer based on insights provided by the sentiment engine. For example, if the user is stressed, the server will provide an answer using gentler language.

[0135] Specific example

[0136] If a user types "What are some ways to deal with the recent rise in interest rates?" on their device and senses anxiety, the server uses an emotion engine to identify the anxiety and provides a reassuring response such as, "If you have a detailed plan regarding interest rate countermeasures, please consult your assigned advisor. They can handle it calmly."

[0137] In this way, communication that takes user emotions into account becomes possible, resulting in a more user-friendly system.

[0138] The following describes the processing flow.

[0139] Step 1:

[0140] The server collects industry-specific data from online databases and APIs. This data includes industry news, regulatory information, and technical documents.

[0141] Step 2:

[0142] The server preprocesses the collected data, removing duplicates, denoising noise, and standardizing the format as needed.

[0143] Step 3:

[0144] The server passes pre-processed data to a machine learning model, which then learns industry-specific knowledge. This model uses natural language processing to understand technical terms and context.

[0145] Step 4:

[0146] The server automatically generates frequently asked questions and their answers using a pre-trained model and stores them in an FAQ database.

[0147] Step 5:

[0148] The user enters a question via their device. For example, they might enter a specific question in natural language, such as, "Tell me about the latest security measures."

[0149] Step 6:

[0150] The terminal sends the entered question to the server. The server searches the FAQ database and identifies the relevant answer.

[0151] Step 7:

[0152] The emotion engine determines the user's emotions from the user's input text and voice. Here, it identifies emotional states such as anxiety, excitement, and anger.

[0153] Step 8:

[0154] The server adjusts the tone and content of the response based on the results of the emotion engine. For example, if the user is nervous, it will generate a response in a calm tone.

[0155] Step 9:

[0156] The server sends the adjusted response to the terminal.

[0157] Step 10:

[0158] The terminal displays the response received from the server to the user. The user can then ask further questions based on this information.

[0159] This processing flow allows users to quickly receive professional and emotionally sensitive responses.

[0160] (Example 2)

[0161] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0162] While conventional AI-based FAQ systems can provide industry-specific information quickly and accurately, they lack the ability to provide personalized responses that consider user emotions. It is necessary to adjust the tone of responses according to the user's feelings to achieve communication that not only provides information but also considers the user's feelings.

[0163] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0164] In this invention, the server includes means for collecting industry-specific information, means for preprocessing the information and training it using a machine learning algorithm, and means for analyzing the user's emotional state and adjusting the representation of the solution according to the user's emotions. This enables not only the provision of information but also a personalized user experience that takes emotions into consideration.

[0165] "Industry-specific information" refers to specialized knowledge and data used within a particular industry, and is generally not widely used in other industries.

[0166] "Preprocessing" is the process of converting raw data into a format that can be used by algorithms, and includes tasks such as denoising, normalization, and data cleaning.

[0167] A "machine learning algorithm" is a computational method that learns patterns from past data and makes predictions and decisions based on new data.

[0168] A "trained algorithm" refers to a machine learning algorithm that has been trained using a specific dataset and has acquired a certain degree of predictive and decision-making ability.

[0169] A "generative AI algorithm" refers to artificial intelligence technology that automatically creates new data or answers based on human instructions.

[0170] "Natural language processing technology" is a technology that enables computers to understand, generate, and translate human language.

[0171] "User emotional state" refers to the psychological state a user is in when seeking information, and includes emotions such as anxiety, excitement, and tension.

[0172] This invention relates to a system that provides personalized responses that take user emotions into account, using an AI system based on industry-specific information. The main components of this system include a server, a terminal, and an emotion recognition engine.

[0173] The server plays a central role in information gathering, model training, and response generation. Specifically, the server collects industry-specific information using remote databases and application programming interfaces (APIs). The collected information is then cleaned and filtered, and used to train a model with machine learning algorithms. This model utilizes natural language processing techniques (such as BERT and GPT) to understand industry jargon and context.

[0174] When a user enters a question via their device, the information is sent to a server, which then searches the FAQ database for relevant answers. During this process, a generative AI algorithm is used to generate new answers even for questions not found in the database.

[0175] The emotion recognition engine analyzes the user's text and voice tone to identify their emotional state. The server uses this insight to adjust the tone of its response before sending it back to the device. For example, if a user makes an anxious input such as, "What should I do about the recent rise in interest rates?", the system will provide a reassuring response such as, "If you have a detailed plan for dealing with interest rates, please consult your advisor. They can handle it calmly."

[0176] As an example of a prompt, one could instruct the AI ​​generation algorithm to "compile a list of frequently asked questions about the industry." This would enable the provision of information that takes into account the user's emotional aspects, resulting in a more user-friendly experience.

[0177] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0178] Step 1:

[0179] The server collects industry-specific information. Specifically, the server retrieves necessary data using remote databases and APIs. The input is information from online sources, and the output is structured data stored on the server's storage. After data retrieval, the server performs data preprocessing such as filtering duplicate data and formatting.

[0180] Step 2:

[0181] The server collects and preprocesses data to train machine learning algorithms. The input is preprocessed industry data, and the output is a trained model capable of understanding industry-specific challenges. This process utilizes natural language processing techniques to improve terminology understanding and context recognition performance.

[0182] Step 3:

[0183] The server uses a pre-trained model to automatically generate frequently asked questions and their answers. It takes the trained model and the prompt "Please compile a list of frequently asked questions about the industry" as input, and the output is a collection of answers stored in an FAQ database. During this process, quality checks are performed to ensure that the automatically generated answers meet specific criteria.

[0184] Step 4:

[0185] The user enters a question into the system via a terminal. The input is text data received from the terminal, and this data is sent to the server. The output is the user's inquiry sent to the server.

[0186] Step 5:

[0187] The server consults an FAQ database and selects the most relevant answer to the user's question. The input is the user's question and the FAQ database, and the output is the most relevant answer. In this process, a generative AI model is used to supplement and generate new answers for questions not found in the database.

[0188] Step 6:

[0189] The emotion recognition engine analyzes the user's emotional state from their input. The input is either text or voice data from the user, and the output is identified emotional information. Based on this information, the server adjusts the tone of its response.

[0190] Step 7:

[0191] The server sends an emotionally balanced response to the device and presents it to the user. The input is the balanced response, and the output is the response displayed to the user on the device. This allows the user to receive information that takes emotional aspects into consideration.

[0192] (Application Example 2)

[0193] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0194] Electronic payment services require prompt and appropriate responses to user concerns and questions during transactions. However, conventional FAQ systems fail to consider the user's emotional state and can only provide standard answers, resulting in insufficient resolution of user anxieties. Furthermore, providing answers that take emotions into account can improve the user experience.

[0195] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0196] In this invention, the server includes means for collecting industry-specific data, means for training a machine learning model, means for automatically generating frequently asked questions and their answers, means for recognizing the user's emotional state and adjusting the tone of the answers accordingly, and means for application in an electronic payment service. As a result, it becomes possible to provide personalized answers that correspond to the user's emotional state, thereby reducing user anxiety and improving their sense of security.

[0197] "Industry-specific data" refers to a collection of information and knowledge related to a particular industry or sector.

[0198] A "machine learning model" is an algorithm that automatically learns patterns and knowledge from data.

[0199] "Frequently asked questions" are common questions that users repeatedly ask.

[0200] "Automatic generation" refers to the process by which a computer creates questions and answers without human intervention.

[0201] "User emotional state" refers to the psychological reactions and feelings of a user at a specific point in time.

[0202] "The tone of the response" refers to the style of language and expression used in the response provided.

[0203] An "electronic payment service" is a digital payment system for transactions conducted over the internet.

[0204] "Personalized responses tailored to emotional state" refer to individually optimized responses that are adjusted based on the user's emotions.

[0205] This invention is for implementing an FAQ system that responds to user emotions in electronic payment services. The main components of the system include a server, a terminal, and an emotion engine.

[0206] The server first collects industry-specific data from online sources. The collected data is preprocessed and trained on a machine learning model. This model uses platforms such as TENSORFLOW® to deepen its understanding of industry-specific terminology and concepts and automatically generate answers to frequently asked questions.

[0207] When a user enters a question via their device, that information is sent to the server. The emotion engine then analyzes the user's input and uses the Google® Cloud Natural Language API to identify their emotional state. For example, if it detects that the user is feeling anxious, it adjusts the tone of the response generated by the server based on that result.

[0208] Ultimately, the server retrieves the most relevant answers from the FAQ database and provides them in language appropriate to the user's emotional state. This allows the user to have a more reassuring experience.

[0209] For example, when a user types "My transfer hasn't been reflected. What should I do?", the server, after having its emotion engine identify the user's anxiety, provides a personalized response in a friendly tone, such as "We understand your concern. It usually takes a few hours, but if it's still not reflected, please contact customer support."

[0210] An example of a prompt for the generating AI model is: "Generate a response to a user who is anxious because their payment has not been reflected, using an appropriate tone based on the user's emotions."

[0211] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0212] Step 1:

[0213] The terminal receives a question from the user. The user enters the question through the terminal's interface, and this input data is sent to the server. The input data is in text format and needs to be properly transmitted over the network for further processing.

[0214] Step 2:

[0215] The server analyzes the user input it receives. The text of the input data is sent to the sentiment engine, which performs sentiment analysis using the Google Cloud Natural Language API. Here, natural language processing techniques are used to identify the user's emotional state (e.g., anxiety, relief, tension) from the input text. The output is data indicating the emotional state.

[0216] Step 3:

[0217] The server searches the FAQ database based on the sentiment analysis results. The server operates a machine learning model (using TensorFlow) to automatically search and extract relevant questions and answers from the FAQ database. In this step, the user's question and analyzed sentiment state are taken as input, and the most appropriate answer candidates are obtained as output.

[0218] Step 4:

[0219] The server adjusts the tone of the response. Using an AI model that generates responses, and referencing the emotional state obtained by the emotion engine, the response tone is harmoniously and appropriately adjusted. Specifically, if the emotion is anxiety, reassuring words and expressions are selected. Here, the input is the emotional state and response data from FAQs, and the output is the final adjusted response.

[0220] Step 5:

[0221] The terminal presents the user with the adjusted response received from the server. The final response is displayed to the user through the terminal's interface, completing the handling of the user's inquiry. This includes data transmission from the server to the terminal, and appropriate feedback is provided to the user as a result of the processing.

[0222] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0223] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0224] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0225] [Second Embodiment]

[0226] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0227] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0228] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0229] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0230] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0231] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0232] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0233] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0234] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0235] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0236] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0237] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. 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".

[0238] This invention is an AI-based FAQ system that provides accurate and rapid answers to frequently asked questions in a specific industry. The system collects industry-specific data and uses machine learning models to train its expertise. It also employs generative AI to generate answers to user questions and maintains and manages the database.

[0239] Data collection and utilization

[0240] The server periodically collects industry-specific information through online databases and APIs. For example, in the medical industry, it would retrieve the latest medical papers and guidelines; in the legal industry, it would retrieve case law and legal amendment information. This ensures that the data is always up-to-date.

[0241] Model training and FAQ generation

[0242] The server preprocesses the collected data and inputs it into a machine learning model. The model learns industry-specific patterns and trends from the collected information and uses this to generate an FAQ database. This automatically creates questions and answers that reflect industry expertise.

[0243] User Interface Support

[0244] Users input questions in natural language from their devices and send them to the server. The server searches its FAQ database for relevant answers and provides them. Even if there is no matching answer in the FAQ database, it uses a generative AI model to supplement it with an appropriate answer. For example, if a user inputs "Tell me about new diabetes treatments" into their device, they can instantly receive a detailed answer based on the latest information.

[0245] This system functions as a specialized question-answering system tailored to specific industries, enabling the rapid and accurate provision of information while minimizing the need for significant human intervention.

[0246] The following describes the processing flow.

[0247] Step 1:

[0248] The server periodically collects industry-specific data from online databases and APIs. This includes collecting the latest news articles, research literature, and regulatory information, and storing it in the database.

[0249] Step 2:

[0250] The server preprocesses the collected data. Specifically, it standardizes the data format, removes noise, and eliminates duplicate information. This ensures that subsequent processing proceeds smoothly.

[0251] Step 3:

[0252] The server uses pre-processed data to input into a machine learning model, allowing it to learn its expertise. This process adjusts the model to understand key patterns and the meaning of technical terms within the data.

[0253] Step 4:

[0254] The server automatically generates FAQs using a pre-trained model. It predicts common questions related to the industry and generates detailed answers to them.

[0255] Step 5:

[0256] The user enters a specific question from their device. For example, they might enter a question in natural language such as, "I want to know about the latest tax law revisions."

[0257] Step 6:

[0258] The terminal formats the user's question appropriately and sends it to the server. The server searches the FAQ database based on the question.

[0259] Step 7:

[0260] The server selects the most relevant answer from the FAQ database and provides it to the user. If no relevant FAQ exists, it uses a generative AI model to generate a new answer.

[0261] Step 8:

[0262] The terminal displays the response received from the server to the user. The user can review the response on the screen and enter further questions if necessary.

[0263] This series of steps allows users to quickly obtain the information they need.

[0264] (Example 1)

[0265] Next, we will describe Example 1. 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."

[0266] Providing quick and accurate answers to frequently asked questions within specific industries is becoming increasingly difficult due to the sheer volume of information and the growing demand for speed. Traditional systems struggle to meet user demands because information updates and the quality of responses are insufficient. Furthermore, the scattered nature of information sources necessitates a means of accurately collecting and efficiently processing relevant information.

[0267] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0268] In this invention, the server includes information aggregation means, means for processing the information and training it using a machine learning algorithm, and means for automatically generating frequently asked questions and their answers using the trained algorithm. This makes it possible to efficiently collect and analyze industry-specific information and provide users with quick and accurate answers.

[0269] "Information aggregation means" refers to methods for automatically acquiring information related to a specific industry through network databases or programmatic interfaces.

[0270] A "machine learning algorithm" is a collection of computer programs designed to learn specific patterns or knowledge from large amounts of data.

[0271] A "trained algorithm" is an algorithm that has been trained by a machine learning algorithm and possesses the ability to generate answers for a specific task.

[0272] A "generative AI algorithm" is an artificial intelligence technology that can generate new information and data based on natural language processing techniques.

[0273] "Information source" refers to a database or knowledge base where answers to user questions are accumulated in the form of FAQs or similar.

[0274] To implement this invention, the server primarily uses information aggregation means, machine learning algorithms, and generative AI algorithms. The server is equipped with information aggregation means for acquiring industry-specific information from online databases and APIs via a network. For example, in the medical industry, the server can periodically collect the latest medical research papers and guidelines.

[0275] The collected information is processed by a preprocessing module on the server. This preprocessing includes tasks such as data cleaning, text normalization, and tokenization. The resulting clean data is then input into a deep learning-based machine learning algorithm to learn industry-specific patterns and trends.

[0276] The trained model is used to quickly generate answers to user questions. Users input questions in natural language through their devices, and the server receives them. The server searches the FAQ database and selects an appropriate answer. If a suitable answer is not found in the database, the server utilizes a generative AI algorithm to generate a new answer. The generative AI uses natural language processing techniques to provide answers in context relevant to the user's question.

[0277] As a specific example, even when a user sends a question from a terminal such as "Tell me about the new diabetes treatment method", the server can immediately provide a detailed answer based on the latest medical information for that request. Examples of prompt sentences include "The user is asking 'Tell me about the new diabetes treatment method'. Please generate a detailed answer based on the latest medical data."

[0278] In this way, the server can provide users with quick and accurate information and propose useful solutions according to the needs of the users.

[0279] The flow of the specific process in Example 1 will be described using FIG. 11.

[0280] Step 1:

[0281] The server uses information aggregation means to obtain industry-specific information from the network. Specifically, the server issues an HTTP request to an online database or API and downloads the latest research papers and guidelines. The input to this process is the URL of the information source or the API endpoint, and the output is saved as raw data.

[0282] Step 2:

[0283] The server preprocesses the obtained raw data. As specific operations, the server executes a data cleaning algorithm to remove noise and inconsistencies. Also, text normalization and tokenization are performed. The input to this step is the raw data obtained in Step 1, and the output is clean data that can be processed by a machine learning algorithm.

[0284] Step 3:

[0285] The server trains a machine learning algorithm using clean data. The server inputs the data into a model to learn patterns and trends in a specific field. The input at this stage is the clean data generated in step 2, and the output is the trained model.

[0286] Step 4:

[0287] The terminal receives a question from the user and sends the question to the server. As a specific operation, the user inputs a question in natural language into the terminal and clicks the "Send" button. The input of this process is the user's question text, and the output is sent as a request to the server.

[0288] Step 5:

[0289] The server first searches the FAQ database for the received question to find a relevant answer. If no relevant answer is found, the server utilizes the generative AI model to create a new answer. The input of this step is the question text obtained in step 4, and the output is the answer from the FAQ database or a new answer generated by the generative AI model.

[0290] Step 6:

[0291] The server provides the generated or retrieved answer to the user. As a specific operation, the server constructs the answer and sends it to the terminal. The input of this process is the answer data in step 5, and the output is the solution displayed on the user terminal.

[0292] (Application Example 1)

[0293] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0294] In physical stores, customers often want to quickly and accurately inquire about product information, sales status, and promotions, but directly asking store staff is often time-consuming. Furthermore, immediate responses are difficult when staff are unavailable. This poses a risk of negatively impacting the customer experience.

[0295] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0296] In this invention, the server includes means for collecting industry-specific information, means for preprocessing the information and training it using a machine learning algorithm, means for automatically generating frequently asked inquiries and their answers using the trained algorithm, and means for customers to query in-store information from a mobile terminal and display the responses on the mobile terminal. This allows customers to quickly obtain necessary information using a mobile device without going through store staff, thereby improving the customer experience.

[0297] "Industry-specific information" refers to information related to a particular industry, and is knowledge and data that is only useful within that industry.

[0298] "Preprocessing" refers to the initial data processing steps performed to prepare raw data into a format that can be easily handled by machine learning algorithms.

[0299] A "machine learning algorithm" refers to a technique that uses data to train a model and then makes predictions or classifications based on new data.

[0300] A "pre-trained algorithm" is a machine learning model that has already been trained and is ready to handle a specific task.

[0301] An "inquiry" refers to a question or request made by a user to obtain information.

[0302] The "Answer Database" is a database that accumulates inquiries generated in the past and answers thereto.

[0303] The "Generative AI Algorithm" is an artificial intelligence technology for generating responses in natural language based on the input data.

[0304] The "Mobile Terminal" refers to portable computer devices such as smartphones and tablets.

[0305] The "In-Store Information" is business information such as products, inventory, promotions, etc. related to a specific physical store.

[0306] The system for implementing this invention consists of a server, a customer's mobile terminal, and a store's information management system. The server collects industry-specific information through an online database or an information-providing interface and preprocesses it. As a result, the data is organized into a form that can be efficiently handled by a machine learning algorithm. The preprocessed data is learned by the server using a machine learning algorithm and utilized for the automatic generation of inquiries and their answers.

[0307] The learned algorithm can accurately understand industry-specific terms and concepts and generate answers using natural language processing technology. When a customer makes an inquiry using a mobile terminal, the server refers to the corresponding inquiry and the answer database to provide a relevant response. If there is no corresponding answer in the database, an answer for supplementation is generated in real time by the generative AI algorithm and displayed on the customer's terminal.

[0308] As a specific example, when a customer inputs "Please tell me about the current special sale in the store" into a mobile terminal in a store, the server retrieves the latest sale information from the database and provides an accurate answer based on it.

[0309] An example of the prompt text for the generative AI model in this case is as follows:

[0310] "Could you tell me about any sales currently running in your store? I'd like to know more about the latest discounts and campaigns."

[0311] This allows customers to quickly and accurately obtain the information they need without having to ask store staff directly.

[0312] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0313] Step 1:

[0314] The server collects industry-specific information through online databases and information provision interfaces. It retrieves information from specified APIs as input and converts unstructured data into an easily understandable format. The output is data that has been prepared for preprocessing.

[0315] Step 2:

[0316] The server preprocesses the collected data. The input is the data prepared in step 1. Data cleaning, tokenization, and feature extraction are performed to transform the data into a format usable by machine learning algorithms. Noise reduction and filtering of necessary data are also performed during this process. The output becomes the training dataset.

[0317] Step 3:

[0318] The server uses pre-processed data to train a machine learning algorithm. It uses a training dataset as input and learns patterns using algorithms such as neural networks. The output is a trained algorithm that can be used to generate queries and answers.

[0319] Step 4:

[0320] A user enters a specific query from a mobile device. The input is natural language text received through the device's interface. The device sends the entered text to the server. The output is the request to the server.

[0321] Step 5:

[0322] The server receives a query from a user and consults its answer database. The input is text sent by the user. The server uses natural language processing techniques to search the database for existing answers related to the query. The output is the answer as a search result.

[0323] Step 6:

[0324] If the answer does not exist in the answer database, the server uses a generative AI model to complete the answer. The input is the query for which no match was found in step 5. The generative AI model uses a prompt to generate a response in natural language. The output is the generated answer text.

[0325] Step 7:

[0326] The server returns the results to the user's terminal. The input is the answer obtained in step 5 or step 6. The terminal displays the received answer on the screen and informs the user. The output is the information displayed to the user.

[0327] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0328] This invention realizes a system that provides more personalized answers by combining an AI-based FAQ system using industry-specific data with a function that recognizes user emotions. This system comprises a server, terminals, and an emotion engine, and provides industry-savvy answers instantly while adjusting the tone and content according to the user's emotional state.

[0329] Data collection and model learning

[0330] The server collects industry-specific data through online databases and APIs. The collected data is preprocessed, and industry expertise is learned using machine learning models. These models utilize natural language processing techniques to understand technical terms and complex concepts.

[0331] FAQ generation and management

[0332] The server uses a pre-trained model to automatically generate frequently asked questions and their answers. The generated FAQs are stored in a database and regularly updated based on the latest information.

[0333] Emotion recognition and response

[0334] When a user enters a question via their device, the emotion engine analyzes the user's wording and, if possible, their tone of voice to identify their emotions. For example, if the user is feeling anxious, the engine adjusts the response to alleviate those emotions.

[0335] User response

[0336] The server selects the most relevant answer from the FAQ database, but adjusts the tone of the answer based on insights provided by the sentiment engine. For example, if the user is stressed, the server will provide an answer using gentler language.

[0337] Specific example

[0338] If a user types "What are some ways to deal with the recent rise in interest rates?" on their device and senses anxiety, the server uses an emotion engine to identify the anxiety and provides a reassuring response such as, "If you have a detailed plan regarding interest rate countermeasures, please consult your assigned advisor. They can handle it calmly."

[0339] In this way, communication that takes user emotions into account becomes possible, resulting in a more user-friendly system.

[0340] The following describes the processing flow.

[0341] Step 1:

[0342] The server collects industry-specific data from online databases and APIs. This data includes industry news, regulatory information, and technical documents.

[0343] Step 2:

[0344] The server preprocesses the collected data, removing duplicates, denoising noise, and standardizing the format as needed.

[0345] Step 3:

[0346] The server passes pre-processed data to a machine learning model, which then learns industry-specific knowledge. This model uses natural language processing to understand technical terms and context.

[0347] Step 4:

[0348] The server automatically generates frequently asked questions and their answers using a pre-trained model and stores them in an FAQ database.

[0349] Step 5:

[0350] The user enters a question via their device. For example, they might enter a specific question in natural language, such as, "Tell me about the latest security measures."

[0351] Step 6:

[0352] The terminal sends the entered question to the server. The server searches the FAQ database and identifies the relevant answer.

[0353] Step 7:

[0354] The emotion engine determines the user's emotions from the user's input text and voice. Here, it identifies emotional states such as anxiety, excitement, and anger.

[0355] Step 8:

[0356] The server adjusts the tone and content of the response based on the results of the emotion engine. For example, if the user is nervous, it will generate a response in a calm tone.

[0357] Step 9:

[0358] The server sends the adjusted response to the terminal.

[0359] Step 10:

[0360] The terminal displays the response received from the server to the user. The user can then ask further questions based on this information.

[0361] This processing flow allows users to quickly receive professional and emotionally sensitive responses.

[0362] (Example 2)

[0363] Next, we will describe Example 2. 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".

[0364] While conventional AI-based FAQ systems can provide industry-specific information quickly and accurately, they lack the ability to provide personalized responses that consider user emotions. It is necessary to adjust the tone of responses according to the user's feelings to achieve communication that not only provides information but also considers the user's feelings.

[0365] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0366] In this invention, the server includes means for collecting industry-specific information, means for preprocessing the information and training it using a machine learning algorithm, and means for analyzing the user's emotional state and adjusting the representation of the solution according to the user's emotions. This enables not only the provision of information but also a personalized user experience that takes emotions into consideration.

[0367] "Industry-specific information" refers to specialized knowledge and data used within a particular industry, and is generally not widely used in other industries.

[0368] "Preprocessing" is the process of converting raw data into a format that can be used by algorithms, and includes tasks such as denoising, normalization, and data cleaning.

[0369] A "machine learning algorithm" is a computational method that learns patterns from past data and makes predictions and decisions based on new data.

[0370] A "trained algorithm" refers to a machine learning algorithm that has been trained using a specific dataset and has acquired a certain degree of predictive and decision-making ability.

[0371] A "generative AI algorithm" refers to artificial intelligence technology that automatically creates new data or answers based on human instructions.

[0372] "Natural language processing technology" is a technology that enables computers to understand, generate, and translate human language.

[0373] "User emotional state" refers to the psychological state a user is in when seeking information, and includes emotions such as anxiety, excitement, and tension.

[0374] This invention relates to a system that provides personalized responses that take user emotions into account, using an AI system based on industry-specific information. The main components of this system include a server, a terminal, and an emotion recognition engine.

[0375] The server plays a central role in information gathering, model training, and response generation. Specifically, the server collects industry-specific information using remote databases and application programming interfaces (APIs). The collected information is then cleaned and filtered, and used to train a model with machine learning algorithms. This model utilizes natural language processing techniques (such as BERT and GPT) to understand industry jargon and context.

[0376] When a user enters a question via their device, the information is sent to a server, which then searches the FAQ database for relevant answers. During this process, a generative AI algorithm is used to generate new answers even for questions not found in the database.

[0377] The emotion recognition engine analyzes the user's text and voice tone to identify their emotional state. The server uses this insight to adjust the tone of its response before sending it back to the device. For example, if a user makes an anxious input such as, "What should I do about the recent rise in interest rates?", the system will provide a reassuring response such as, "If you have a detailed plan for dealing with interest rates, please consult your advisor. They can handle it calmly."

[0378] As an example of a prompt, one could instruct the AI ​​generation algorithm to "compile a list of frequently asked questions about the industry." This would enable the provision of information that takes into account the user's emotional aspects, resulting in a more user-friendly experience.

[0379] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0380] Step 1:

[0381] The server collects industry-specific information. Specifically, the server retrieves necessary data using remote databases and APIs. The input is information from online sources, and the output is structured data stored on the server's storage. After data retrieval, the server performs data preprocessing such as filtering duplicate data and formatting.

[0382] Step 2:

[0383] The server collects and preprocesses data to train machine learning algorithms. The input is preprocessed industry data, and the output is a trained model capable of understanding industry-specific challenges. This process utilizes natural language processing techniques to improve terminology understanding and context recognition performance.

[0384] Step 3:

[0385] The server uses a pre-trained model to automatically generate frequently asked questions and their answers. It takes the trained model and the prompt "Please compile a list of frequently asked questions about the industry" as input, and the output is a collection of answers stored in an FAQ database. During this process, quality checks are performed to ensure that the automatically generated answers meet specific criteria.

[0386] Step 4:

[0387] The user enters a question into the system via a terminal. The input is text data received from the terminal, and this data is sent to the server. The output is the user's inquiry sent to the server.

[0388] Step 5:

[0389] The server consults an FAQ database and selects the most relevant answer to the user's question. The input is the user's question and the FAQ database, and the output is the most relevant answer. In this process, a generative AI model is used to supplement and generate new answers for questions not found in the database.

[0390] Step 6:

[0391] The emotion recognition engine analyzes the user's emotional state from their input. The input is either text or voice data from the user, and the output is identified emotional information. Based on this information, the server adjusts the tone of its response.

[0392] Step 7:

[0393] The server sends an emotionally balanced response to the device and presents it to the user. The input is the balanced response, and the output is the response displayed to the user on the device. This allows the user to receive information that takes emotional aspects into consideration.

[0394] (Application Example 2)

[0395] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0396] Electronic payment services require prompt and appropriate responses to user concerns and questions during transactions. However, conventional FAQ systems fail to consider the user's emotional state and can only provide standard answers, resulting in insufficient resolution of user anxieties. Furthermore, providing answers that take emotions into account can improve the user experience.

[0397] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0398] In this invention, the server includes means for collecting industry-specific data, means for training a machine learning model, means for automatically generating frequently asked questions and their answers, means for recognizing the user's emotional state and adjusting the tone of the answers accordingly, and means for application in an electronic payment service. As a result, it becomes possible to provide personalized answers that correspond to the user's emotional state, thereby reducing user anxiety and improving their sense of security.

[0399] "Industry-specific data" refers to a collection of information and knowledge related to a particular industry or sector.

[0400] A "machine learning model" is an algorithm that automatically learns patterns and knowledge from data.

[0401] "Frequently asked questions" are common questions that users repeatedly ask.

[0402] "Automatic generation" refers to the process by which a computer creates questions and answers without human intervention.

[0403] "User emotional state" refers to the psychological reactions and feelings of a user at a specific point in time.

[0404] "The tone of the response" refers to the style of language and expression used in the response provided.

[0405] An "electronic payment service" is a digital payment system for transactions conducted over the internet.

[0406] "Personalized responses tailored to emotional state" refer to individually optimized responses that are adjusted based on the user's emotions.

[0407] This invention is for implementing an FAQ system that responds to user emotions in electronic payment services. The main components of the system include a server, a terminal, and an emotion engine.

[0408] The server first collects industry-specific data from online sources. The collected data is preprocessed and trained on a machine learning model. This model uses platforms such as TensorFlow to deepen its understanding of industry-specific terminology and concepts and automatically generate answers to frequently asked questions.

[0409] When a user enters a question via their device, that information is sent to the server. The emotion engine then analyzes the user's input and uses the Google Cloud Natural Language API to identify their emotional state. For example, if it detects that the user is feeling anxious, it adjusts the tone of the response generated by the server based on that result.

[0410] Ultimately, the server retrieves the most relevant answers from the FAQ database and provides them in language appropriate to the user's emotional state. This allows the user to have a more reassuring experience.

[0411] For example, when a user types "My transfer hasn't been reflected. What should I do?", the server, after having its emotion engine identify the user's anxiety, provides a personalized response in a friendly tone, such as "We understand your concern. It usually takes a few hours, but if it's still not reflected, please contact customer support."

[0412] An example of a prompt for the generating AI model is: "Generate a response to a user who is anxious because their payment has not been reflected, using an appropriate tone based on the user's emotions."

[0413] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0414] Step 1:

[0415] The terminal receives a question from the user. The user enters the question through the terminal's interface, and this input data is sent to the server. The input data is in text format and needs to be properly transmitted over the network for further processing.

[0416] Step 2:

[0417] The server analyzes the user input it receives. The text of the input data is sent to the sentiment engine, which performs sentiment analysis using the Google Cloud Natural Language API. Here, natural language processing techniques are used to identify the user's emotional state (e.g., anxiety, relief, tension) from the input text. The output is data indicating the emotional state.

[0418] Step 3:

[0419] The server searches the FAQ database based on the sentiment analysis results. The server operates a machine learning model (using TensorFlow) to automatically search and extract relevant questions and answers from the FAQ database. In this step, the user's question and analyzed sentiment state are taken as input, and the most appropriate answer candidates are obtained as output.

[0420] Step 4:

[0421] The server adjusts the tone of the response. Using an AI model that generates responses, and referencing the emotional state obtained by the emotion engine, the response tone is harmoniously and appropriately adjusted. Specifically, if the emotion is anxiety, reassuring words and expressions are selected. Here, the input is the emotional state and response data from FAQs, and the output is the final adjusted response.

[0422] Step 5:

[0423] The terminal presents the user with the adjusted response received from the server. The final response is displayed to the user through the terminal's interface, completing the handling of the user's inquiry. This includes data transmission from the server to the terminal, and appropriate feedback is provided to the user as a result of the processing.

[0424] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0425] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0426] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0427] [Third Embodiment]

[0428] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0429] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0430] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0431] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0432] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0433] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0434] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0435] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0436] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0437] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0438] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0439] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0440] This invention is an AI-based FAQ system that provides accurate and rapid answers to frequently asked questions in a specific industry. The system collects industry-specific data and uses machine learning models to train its expertise. It also employs generative AI to generate answers to user questions and maintains and manages the database.

[0441] Data collection and utilization

[0442] The server periodically collects industry-specific information through online databases and APIs. For example, in the medical industry, it would retrieve the latest medical papers and guidelines; in the legal industry, it would retrieve case law and legal amendment information. This ensures that the data is always up-to-date.

[0443] Model training and FAQ generation

[0444] The server preprocesses the collected data and inputs it into a machine learning model. The model learns industry-specific patterns and trends from the collected information and uses this to generate an FAQ database. This automatically creates questions and answers that reflect industry expertise.

[0445] User Interface Support

[0446] Users input questions in natural language from their devices and send them to the server. The server searches its FAQ database for relevant answers and provides them. Even if there is no matching answer in the FAQ database, it uses a generative AI model to supplement it with an appropriate answer. For example, if a user inputs "Tell me about new diabetes treatments" into their device, they can instantly receive a detailed answer based on the latest information.

[0447] This system functions as a specialized question-answering system tailored to specific industries, enabling the rapid and accurate provision of information while minimizing the need for significant human intervention.

[0448] The following describes the processing flow.

[0449] Step 1:

[0450] The server periodically collects industry-specific data from online databases and APIs. This includes collecting the latest news articles, research literature, and regulatory information, and storing it in the database.

[0451] Step 2:

[0452] The server preprocesses the collected data. Specifically, it standardizes the data format, removes noise, and eliminates duplicate information. This ensures that subsequent processing proceeds smoothly.

[0453] Step 3:

[0454] The server uses pre-processed data to input into a machine learning model, allowing it to learn its expertise. This process adjusts the model to understand key patterns and the meaning of technical terms within the data.

[0455] Step 4:

[0456] The server automatically generates FAQs using a pre-trained model. It predicts common questions related to the industry and generates detailed answers to them.

[0457] Step 5:

[0458] The user enters a specific question from their device. For example, they might enter a question in natural language such as, "I want to know about the latest tax law revisions."

[0459] Step 6:

[0460] The terminal formats the user's question appropriately and sends it to the server. The server searches the FAQ database based on the question.

[0461] Step 7:

[0462] The server selects the most relevant answer from the FAQ database and provides it to the user. If no relevant FAQ exists, it uses a generative AI model to generate a new answer.

[0463] Step 8:

[0464] The terminal displays the response received from the server to the user. The user can review the response on the screen and enter further questions if necessary.

[0465] This series of steps allows users to quickly obtain the information they need.

[0466] (Example 1)

[0467] Next, we will describe Example 1. 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."

[0468] Providing quick and accurate answers to frequently asked questions within specific industries is becoming increasingly difficult due to the sheer volume of information and the growing demand for speed. Traditional systems struggle to meet user demands because information updates and the quality of responses are insufficient. Furthermore, the scattered nature of information sources necessitates a means of accurately collecting and efficiently processing relevant information.

[0469] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0470] In this invention, the server includes information aggregation means, means for processing the information and training it using a machine learning algorithm, and means for automatically generating frequently asked questions and their answers using the trained algorithm. This makes it possible to efficiently collect and analyze industry-specific information and provide users with quick and accurate answers.

[0471] "Information aggregation means" refers to methods for automatically acquiring information related to a specific industry through network databases or programmatic interfaces.

[0472] A "machine learning algorithm" is a collection of computer programs designed to learn specific patterns or knowledge from large amounts of data.

[0473] A "trained algorithm" is an algorithm that has been trained by a machine learning algorithm and possesses the ability to generate answers for a specific task.

[0474] A "generative AI algorithm" is an artificial intelligence technology that can generate new information and data based on natural language processing techniques.

[0475] "Information source" refers to a database or knowledge base where answers to user questions are accumulated in the form of FAQs or similar.

[0476] To implement this invention, the server primarily uses information aggregation means, machine learning algorithms, and generative AI algorithms. The server is equipped with information aggregation means for acquiring industry-specific information from online databases and APIs via a network. For example, in the medical industry, the server can periodically collect the latest medical research papers and guidelines.

[0477] The collected information is processed by a preprocessing module on the server. This preprocessing includes tasks such as data cleaning, text normalization, and tokenization. The resulting clean data is then input into a deep learning-based machine learning algorithm to learn industry-specific patterns and trends.

[0478] The trained model is used to quickly generate answers to user questions. Users input questions in natural language through their devices, and the server receives them. The server searches the FAQ database and selects an appropriate answer. If a suitable answer is not found in the database, the server utilizes a generative AI algorithm to generate a new answer. The generative AI uses natural language processing techniques to provide answers in context relevant to the user's question.

[0479] For example, even if a user sends a question from their device such as "Tell me about new diabetes treatments," the server can immediately provide a detailed answer based on the latest medical information. An example of a prompt message would be, "The user is asking 'Tell me about new diabetes treatments.' Please generate a detailed answer based on the latest medical data."

[0480] In this way, the server can provide users with quick and accurate information and propose useful solutions tailored to their needs.

[0481] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0482] Step 1:

[0483] The server uses information aggregation tools to retrieve industry-specific information from the network. Specifically, the server issues HTTP requests to online databases and APIs to download the latest research papers and guidelines. The input to this process is the URL of the information source or the API endpoint, and the output is stored as raw data.

[0484] Step 2:

[0485] The server preprocesses the acquired raw data. Specifically, the server runs a data cleaning algorithm to remove noise and inconsistencies. Text normalization and tokenization are also performed. The input to this process is the raw data obtained in step 1, and the output is clean data that can be processed by a machine learning algorithm.

[0486] Step 3:

[0487] The server trains a machine learning algorithm using clean data. The server inputs the data into the model, allowing it to learn patterns and trends in a specific domain. At this stage, the input is the clean data generated in step 2, and the output is the trained model.

[0488] Step 4:

[0489] The terminal receives a question from the user and sends that question to the server. Specifically, the user enters the question in natural language into the terminal and clicks the "Send" button. The input in this process is the user's question, and the output is sent as a request to the server.

[0490] Step 5:

[0491] The server first searches the FAQ database for the received question to find a relevant answer. If no relevant answer is found, the server uses a generative AI model to create a new answer. The input for this step is the question obtained in step 4, and the output is either an answer from the FAQ database or a new answer from the generative AI model.

[0492] Step 6:

[0493] The server provides the user with the generated or retrieved answer. Specifically, the server composes the answer and sends it to the terminal. The input to this process is the answer data from step 5, and the output is the answer displayed on the user's terminal.

[0494] (Application Example 1)

[0495] Next, we will explain Application Example 1. In the following explanation, 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."

[0496] In physical stores, customers often want to quickly and accurately inquire about product information, sales status, and promotions, but directly asking store staff is often time-consuming. Furthermore, immediate responses are difficult when staff are unavailable. This poses a risk of negatively impacting the customer experience.

[0497] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0498] In this invention, the server includes means for collecting industry-specific information, means for preprocessing the information and training it using a machine learning algorithm, means for automatically generating frequently asked inquiries and their answers using the trained algorithm, and means for customers to query in-store information from a mobile terminal and display the responses on the mobile terminal. This allows customers to quickly obtain necessary information using a mobile device without going through store staff, thereby improving the customer experience.

[0499] "Industry-specific information" refers to information related to a particular industry, and is knowledge and data that is only useful within that industry.

[0500] "Preprocessing" refers to the initial data processing steps performed to prepare raw data into a format that can be easily handled by machine learning algorithms.

[0501] A "machine learning algorithm" refers to a technique that uses data to train a model and then makes predictions or classifications based on new data.

[0502] A "pre-trained algorithm" is a machine learning model that has already been trained and is ready to handle a specific task.

[0503] An "inquiry" refers to a question or request made by a user to obtain information.

[0504] The "Answer Information Repository" is a database that stores past inquiries and their corresponding answers.

[0505] A "generative AI algorithm" is an artificial intelligence technology that generates natural language responses based on input data.

[0506] "Mobile devices" refer to portable computer devices such as smartphones and tablets.

[0507] "In-store information" refers to business information related to a specific physical store, such as products, inventory, and promotions.

[0508] The system for implementing this invention consists of a server, a customer's mobile terminal, and a store information management system. The server collects industry-specific information through online databases and information provision interfaces and preprocesses it. This prepares the data in a format that can be efficiently handled by machine learning algorithms. The preprocessed data is then trained by the server using machine learning algorithms and used to automatically generate inquiries and their answers.

[0509] The trained algorithm uses natural language processing techniques to accurately understand industry-specific terminology and concepts and generate responses. When a customer makes a query using a mobile device, the server refers to the relevant query and answer database to provide a corresponding response. If a matching answer does not exist in the database, the generation AI algorithm generates a supplementary answer in real time and displays it on the customer's device.

[0510] For example, if a customer in a store enters "Please tell me about the current special sale in the store" into a mobile terminal, the server retrieves the latest sale information from its database and provides an accurate response based on that information.

[0511] An example of a prompt for the generative AI model in this case is as follows:

[0512] "Could you tell me about any sales currently running in your store? I'd like to know more about the latest discounts and campaigns."

[0513] This allows customers to quickly and accurately obtain the information they need without having to ask store staff directly.

[0514] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0515] Step 1:

[0516] The server collects industry-specific information through online databases and information provision interfaces. It retrieves information from specified APIs as input and converts unstructured data into an easily understandable format. The output is data that has been prepared for preprocessing.

[0517] Step 2:

[0518] The server preprocesses the collected data. The input is the data prepared in step 1. Data cleaning, tokenization, and feature extraction are performed to transform the data into a format usable by machine learning algorithms. Noise reduction and filtering of necessary data are also performed during this process. The output becomes the training dataset.

[0519] Step 3:

[0520] The server uses pre-processed data to train a machine learning algorithm. It uses a training dataset as input and learns patterns using algorithms such as neural networks. The output is a trained algorithm that can be used to generate queries and answers.

[0521] Step 4:

[0522] A user enters a specific query from a mobile device. The input is natural language text received through the device's interface. The device sends the entered text to the server. The output is the request to the server.

[0523] Step 5:

[0524] The server receives a query from a user and consults its answer database. The input is text sent by the user. The server uses natural language processing techniques to search the database for existing answers related to the query. The output is the answer as a search result.

[0525] Step 6:

[0526] If the answer does not exist in the answer database, the server uses a generative AI model to complete the answer. The input is the query for which no match was found in step 5. The generative AI model uses a prompt to generate a response in natural language. The output is the generated answer text.

[0527] Step 7:

[0528] The server returns the results to the user's terminal. The input is the answer obtained in step 5 or step 6. The terminal displays the received answer on the screen and informs the user. The output is the information displayed to the user.

[0529] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0530] This invention realizes a system that provides more personalized answers by combining an AI-based FAQ system using industry-specific data with a function that recognizes user emotions. This system comprises a server, terminals, and an emotion engine, and provides industry-savvy answers instantly while adjusting the tone and content according to the user's emotional state.

[0531] Data collection and model learning

[0532] The server collects industry-specific data through online databases and APIs. The collected data is preprocessed, and industry expertise is learned using machine learning models. These models utilize natural language processing techniques to understand technical terms and complex concepts.

[0533] FAQ generation and management

[0534] The server uses a pre-trained model to automatically generate frequently asked questions and their answers. The generated FAQs are stored in a database and regularly updated based on the latest information.

[0535] Emotion recognition and response

[0536] When a user enters a question via their device, the emotion engine analyzes the user's wording and, if possible, their tone of voice to identify their emotions. For example, if the user is feeling anxious, the engine adjusts the response to alleviate those emotions.

[0537] User response

[0538] The server selects the most relevant answer from the FAQ database, but adjusts the tone of the answer based on insights provided by the sentiment engine. For example, if the user is stressed, the server will provide an answer using gentler language.

[0539] Specific example

[0540] If a user types "What are some ways to deal with the recent rise in interest rates?" on their device and senses anxiety, the server uses an emotion engine to identify the anxiety and provides a reassuring response such as, "If you have a detailed plan regarding interest rate countermeasures, please consult your assigned advisor. They can handle it calmly."

[0541] In this way, communication that takes user emotions into account becomes possible, resulting in a more user-friendly system.

[0542] The following describes the processing flow.

[0543] Step 1:

[0544] The server collects industry-specific data from online databases and APIs. This data includes industry news, regulatory information, and technical documents.

[0545] Step 2:

[0546] The server preprocesses the collected data, removing duplicates, denoising noise, and standardizing the format as needed.

[0547] Step 3:

[0548] The server passes pre-processed data to a machine learning model, which then learns industry-specific knowledge. This model uses natural language processing to understand technical terms and context.

[0549] Step 4:

[0550] The server automatically generates frequently asked questions and their answers using a pre-trained model and stores them in an FAQ database.

[0551] Step 5:

[0552] The user enters a question via their device. For example, they might enter a specific question in natural language, such as, "Tell me about the latest security measures."

[0553] Step 6:

[0554] The terminal sends the entered question to the server. The server searches the FAQ database and identifies the relevant answer.

[0555] Step 7:

[0556] The emotion engine determines the user's emotions from the user's input text and voice. Here, it identifies emotional states such as anxiety, excitement, and anger.

[0557] Step 8:

[0558] The server adjusts the tone and content of the response based on the results of the emotion engine. For example, if the user is nervous, it will generate a response in a calm tone.

[0559] Step 9:

[0560] The server sends the adjusted response to the terminal.

[0561] Step 10:

[0562] The terminal displays the response received from the server to the user. The user can then ask further questions based on this information.

[0563] This processing flow allows users to quickly receive professional and emotionally sensitive responses.

[0564] (Example 2)

[0565] Next, we will describe Example 2. 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."

[0566] While conventional AI-based FAQ systems can provide industry-specific information quickly and accurately, they lack the ability to provide personalized responses that consider user emotions. It is necessary to adjust the tone of responses according to the user's feelings to achieve communication that not only provides information but also considers the user's feelings.

[0567] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0568] In this invention, the server includes means for collecting industry-specific information, means for preprocessing the information and training it using a machine learning algorithm, and means for analyzing the user's emotional state and adjusting the representation of the solution according to the user's emotions. This enables not only the provision of information but also a personalized user experience that takes emotions into consideration.

[0569] "Industry-specific information" refers to specialized knowledge and data used within a particular industry, and is generally not widely used in other industries.

[0570] "Preprocessing" is the process of converting raw data into a format that can be used by algorithms, and includes tasks such as denoising, normalization, and data cleaning.

[0571] A "machine learning algorithm" is a computational method that learns patterns from past data and makes predictions and decisions based on new data.

[0572] A "trained algorithm" refers to a machine learning algorithm that has been trained using a specific dataset and has acquired a certain degree of predictive and decision-making ability.

[0573] A "generative AI algorithm" refers to artificial intelligence technology that automatically creates new data or answers based on human instructions.

[0574] "Natural language processing technology" is a technology that enables computers to understand, generate, and translate human language.

[0575] "User emotional state" refers to the psychological state a user is in when seeking information, and includes emotions such as anxiety, excitement, and tension.

[0576] This invention relates to a system that provides personalized responses that take user emotions into account, using an AI system based on industry-specific information. The main components of this system include a server, a terminal, and an emotion recognition engine.

[0577] The server plays a central role in information gathering, model training, and response generation. Specifically, the server collects industry-specific information using remote databases and application programming interfaces (APIs). The collected information is then cleaned and filtered, and used to train a model with machine learning algorithms. This model utilizes natural language processing techniques (such as BERT and GPT) to understand industry jargon and context.

[0578] When a user enters a question via their device, the information is sent to a server, which then searches the FAQ database for relevant answers. During this process, a generative AI algorithm is used to generate new answers even for questions not found in the database.

[0579] The emotion recognition engine analyzes the user's text and voice tone to identify their emotional state. The server uses this insight to adjust the tone of its response before sending it back to the device. For example, if a user makes an anxious input such as, "What should I do about the recent rise in interest rates?", the system will provide a reassuring response such as, "If you have a detailed plan for dealing with interest rates, please consult your advisor. They can handle it calmly."

[0580] As an example of a prompt, one could instruct the AI ​​generation algorithm to "compile a list of frequently asked questions about the industry." This would enable the provision of information that takes into account the user's emotional aspects, resulting in a more user-friendly experience.

[0581] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0582] Step 1:

[0583] The server collects industry-specific information. Specifically, the server retrieves necessary data using remote databases and APIs. The input is information from online sources, and the output is structured data stored on the server's storage. After data retrieval, the server performs data preprocessing such as filtering duplicate data and formatting.

[0584] Step 2:

[0585] The server collects and preprocesses data to train machine learning algorithms. The input is preprocessed industry data, and the output is a trained model capable of understanding industry-specific challenges. This process utilizes natural language processing techniques to improve terminology understanding and context recognition performance.

[0586] Step 3:

[0587] The server uses a pre-trained model to automatically generate frequently asked questions and their answers. It takes the trained model and the prompt "Please compile a list of frequently asked questions about the industry" as input, and the output is a collection of answers stored in an FAQ database. During this process, quality checks are performed to ensure that the automatically generated answers meet specific criteria.

[0588] Step 4:

[0589] The user enters a question into the system via a terminal. The input is text data received from the terminal, and this data is sent to the server. The output is the user's inquiry sent to the server.

[0590] Step 5:

[0591] The server consults an FAQ database and selects the most relevant answer to the user's question. The input is the user's question and the FAQ database, and the output is the most relevant answer. In this process, a generative AI model is used to supplement and generate new answers for questions not found in the database.

[0592] Step 6:

[0593] The emotion recognition engine analyzes the user's emotional state from their input. The input is either text or voice data from the user, and the output is identified emotional information. Based on this information, the server adjusts the tone of its response.

[0594] Step 7:

[0595] The server sends an emotionally balanced response to the device and presents it to the user. The input is the balanced response, and the output is the response displayed to the user on the device. This allows the user to receive information that takes emotional aspects into consideration.

[0596] (Application Example 2)

[0597] Next, we will explain application example 2. In the following explanation, 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."

[0598] Electronic payment services require prompt and appropriate responses to user concerns and questions during transactions. However, conventional FAQ systems fail to consider the user's emotional state and can only provide standard answers, resulting in insufficient resolution of user anxieties. Furthermore, providing answers that take emotions into account can improve the user experience.

[0599] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0600] In this invention, the server includes means for collecting industry-specific data, means for training a machine learning model, means for automatically generating frequently asked questions and their answers, means for recognizing the user's emotional state and adjusting the tone of the answers accordingly, and means for application in an electronic payment service. As a result, it becomes possible to provide personalized answers that correspond to the user's emotional state, thereby reducing user anxiety and improving their sense of security.

[0601] "Industry-specific data" refers to a collection of information and knowledge related to a particular industry or sector.

[0602] A "machine learning model" is an algorithm that automatically learns patterns and knowledge from data.

[0603] "Frequently asked questions" are common questions that users repeatedly ask.

[0604] "Automatic generation" refers to the process by which a computer creates questions and answers without human intervention.

[0605] "User emotional state" refers to the psychological reactions and feelings of a user at a specific point in time.

[0606] "The tone of the response" refers to the style of language and expression used in the response provided.

[0607] An "electronic payment service" is a digital payment system for transactions conducted over the internet.

[0608] "Personalized responses tailored to emotional state" refer to individually optimized responses that are adjusted based on the user's emotions.

[0609] This invention is for implementing an FAQ system that responds to user emotions in electronic payment services. The main components of the system include a server, a terminal, and an emotion engine.

[0610] The server first collects industry-specific data from online sources. The collected data is preprocessed and trained on a machine learning model. This model uses platforms such as TensorFlow to deepen its understanding of industry-specific terminology and concepts and automatically generate answers to frequently asked questions.

[0611] When a user enters a question via their device, that information is sent to the server. The emotion engine then analyzes the user's input and uses the Google Cloud Natural Language API to identify their emotional state. For example, if it detects that the user is feeling anxious, it adjusts the tone of the response generated by the server based on that result.

[0612] Ultimately, the server retrieves the most relevant answers from the FAQ database and provides them in language appropriate to the user's emotional state. This allows the user to have a more reassuring experience.

[0613] For example, when a user types "My transfer hasn't been reflected. What should I do?", the server, after having its emotion engine identify the user's anxiety, provides a personalized response in a friendly tone, such as "We understand your concern. It usually takes a few hours, but if it's still not reflected, please contact customer support."

[0614] An example of a prompt for the generating AI model is: "Generate a response to a user who is anxious because their payment has not been reflected, using an appropriate tone based on the user's emotions."

[0615] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0616] Step 1:

[0617] The terminal receives a question from the user. The user enters the question through the terminal's interface, and this input data is sent to the server. The input data is in text format and needs to be properly transmitted over the network for further processing.

[0618] Step 2:

[0619] The server analyzes the user input it receives. The text of the input data is sent to the sentiment engine, which performs sentiment analysis using the Google Cloud Natural Language API. Here, natural language processing techniques are used to identify the user's emotional state (e.g., anxiety, relief, tension) from the input text. The output is data indicating the emotional state.

[0620] Step 3:

[0621] The server searches the FAQ database based on the sentiment analysis results. The server operates a machine learning model (using TensorFlow) to automatically search and extract relevant questions and answers from the FAQ database. In this step, the user's question and analyzed sentiment state are taken as input, and the most appropriate answer candidates are obtained as output.

[0622] Step 4:

[0623] The server adjusts the tone of the response. Using an AI model that generates responses, and referencing the emotional state obtained by the emotion engine, the response tone is harmoniously and appropriately adjusted. Specifically, if the emotion is anxiety, reassuring words and expressions are selected. Here, the input is the emotional state and response data from FAQs, and the output is the final adjusted response.

[0624] Step 5:

[0625] The terminal presents the user with the adjusted response received from the server. The final response is displayed to the user through the terminal's interface, completing the handling of the user's inquiry. This includes data transmission from the server to the terminal, and appropriate feedback is provided to the user as a result of the processing.

[0626] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0627] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0628] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0629] [Fourth Embodiment]

[0630] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0631] As shown in Figure 7, the 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.

[0632] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0633] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0634] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0635] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0636] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0637] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0638] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0639] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0640] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0641] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0642] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0643] This invention is an AI-based FAQ system that provides accurate and rapid answers to frequently asked questions in a specific industry. The system collects industry-specific data and uses machine learning models to train its expertise. It also employs generative AI to generate answers to user questions and maintains and manages the database.

[0644] Data collection and utilization

[0645] The server periodically collects industry-specific information through online databases and APIs. For example, in the medical industry, it would retrieve the latest medical papers and guidelines; in the legal industry, it would retrieve case law and legal amendment information. This ensures that the data is always up-to-date.

[0646] Model training and FAQ generation

[0647] The server preprocesses the collected data and inputs it into a machine learning model. The model learns industry-specific patterns and trends from the collected information and uses this to generate an FAQ database. This automatically creates questions and answers that reflect industry expertise.

[0648] User Interface Support

[0649] Users input questions in natural language from their devices and send them to the server. The server searches its FAQ database for relevant answers and provides them. Even if there is no matching answer in the FAQ database, it uses a generative AI model to supplement it with an appropriate answer. For example, if a user inputs "Tell me about new diabetes treatments" into their device, they can instantly receive a detailed answer based on the latest information.

[0650] This system functions as a specialized question-answering system tailored to specific industries, enabling the rapid and accurate provision of information while minimizing the need for significant human intervention.

[0651] The following describes the processing flow.

[0652] Step 1:

[0653] The server periodically collects industry-specific data from online databases and APIs. This includes collecting the latest news articles, research literature, and regulatory information, and storing it in the database.

[0654] Step 2:

[0655] The server preprocesses the collected data. Specifically, it standardizes the data format, removes noise, and eliminates duplicate information. This ensures that subsequent processing proceeds smoothly.

[0656] Step 3:

[0657] The server uses pre-processed data to input into a machine learning model, allowing it to learn its expertise. This process adjusts the model to understand key patterns and the meaning of technical terms within the data.

[0658] Step 4:

[0659] The server automatically generates FAQs using a pre-trained model. It predicts common questions related to the industry and generates detailed answers to them.

[0660] Step 5:

[0661] The user enters a specific question from their device. For example, they might enter a question in natural language such as, "I want to know about the latest tax law revisions."

[0662] Step 6:

[0663] The terminal formats the user's question appropriately and sends it to the server. The server searches the FAQ database based on the question.

[0664] Step 7:

[0665] The server selects the most relevant answer from the FAQ database and provides it to the user. If no relevant FAQ exists, it uses a generative AI model to generate a new answer.

[0666] Step 8:

[0667] The terminal displays the response received from the server to the user. The user can review the response on the screen and enter further questions if necessary.

[0668] This series of steps allows users to quickly obtain the information they need.

[0669] (Example 1)

[0670] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0671] Providing quick and accurate answers to frequently asked questions within specific industries is becoming increasingly difficult due to the sheer volume of information and the growing demand for speed. Traditional systems struggle to meet user demands because information updates and the quality of responses are insufficient. Furthermore, the scattered nature of information sources necessitates a means of accurately collecting and efficiently processing relevant information.

[0672] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0673] In this invention, the server includes information aggregation means, means for processing the information and training it using a machine learning algorithm, and means for automatically generating frequently asked questions and their answers using the trained algorithm. This makes it possible to efficiently collect and analyze industry-specific information and provide users with quick and accurate answers.

[0674] "Information aggregation means" refers to methods for automatically acquiring information related to a specific industry through network databases or programmatic interfaces.

[0675] A "machine learning algorithm" is a collection of computer programs designed to learn specific patterns or knowledge from large amounts of data.

[0676] A "trained algorithm" is an algorithm that has been trained by a machine learning algorithm and possesses the ability to generate answers for a specific task.

[0677] A "generative AI algorithm" is an artificial intelligence technology that can generate new information and data based on natural language processing techniques.

[0678] "Information source" refers to a database or knowledge base where answers to user questions are accumulated in the form of FAQs or similar.

[0679] To implement this invention, the server primarily uses information aggregation means, machine learning algorithms, and generative AI algorithms. The server is equipped with information aggregation means for acquiring industry-specific information from online databases and APIs via a network. For example, in the medical industry, the server can periodically collect the latest medical research papers and guidelines.

[0680] The collected information is processed by a preprocessing module on the server. This preprocessing includes tasks such as data cleaning, text normalization, and tokenization. The resulting clean data is then input into a deep learning-based machine learning algorithm to learn industry-specific patterns and trends.

[0681] The trained model is used to quickly generate answers to user questions. Users input questions in natural language through their devices, and the server receives them. The server searches the FAQ database and selects an appropriate answer. If a suitable answer is not found in the database, the server utilizes a generative AI algorithm to generate a new answer. The generative AI uses natural language processing techniques to provide answers in context relevant to the user's question.

[0682] For example, even if a user sends a question from their device such as "Tell me about new diabetes treatments," the server can immediately provide a detailed answer based on the latest medical information. An example of a prompt message would be, "The user is asking 'Tell me about new diabetes treatments.' Please generate a detailed answer based on the latest medical data."

[0683] In this way, the server can provide users with quick and accurate information and propose useful solutions tailored to their needs.

[0684] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0685] Step 1:

[0686] The server uses information aggregation tools to retrieve industry-specific information from the network. Specifically, the server issues HTTP requests to online databases and APIs to download the latest research papers and guidelines. The input to this process is the URL of the information source or the API endpoint, and the output is stored as raw data.

[0687] Step 2:

[0688] The server preprocesses the acquired raw data. Specifically, the server runs a data cleaning algorithm to remove noise and inconsistencies. Text normalization and tokenization are also performed. The input to this process is the raw data obtained in step 1, and the output is clean data that can be processed by a machine learning algorithm.

[0689] Step 3:

[0690] The server trains a machine learning algorithm using clean data. The server inputs the data into the model, allowing it to learn patterns and trends in a specific domain. At this stage, the input is the clean data generated in step 2, and the output is the trained model.

[0691] Step 4:

[0692] The terminal receives a question from the user and sends that question to the server. Specifically, the user enters the question in natural language into the terminal and clicks the "Send" button. The input in this process is the user's question, and the output is sent as a request to the server.

[0693] Step 5:

[0694] The server first searches the FAQ database for the received question to find a relevant answer. If no relevant answer is found, the server uses a generative AI model to create a new answer. The input for this step is the question obtained in step 4, and the output is either an answer from the FAQ database or a new answer from the generative AI model.

[0695] Step 6:

[0696] The server provides the user with the generated or retrieved answer. Specifically, the server composes the answer and sends it to the terminal. The input to this process is the answer data from step 5, and the output is the answer displayed on the user's terminal.

[0697] (Application Example 1)

[0698] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0699] In physical stores, customers often want to quickly and accurately inquire about product information, sales status, and promotions, but directly asking store staff is often time-consuming. Furthermore, immediate responses are difficult when staff are unavailable. This poses a risk of negatively impacting the customer experience.

[0700] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0701] In this invention, the server includes means for collecting industry-specific information, means for preprocessing the information and training it using a machine learning algorithm, means for automatically generating frequently asked inquiries and their answers using the trained algorithm, and means for customers to query in-store information from a mobile terminal and display the responses on the mobile terminal. This allows customers to quickly obtain necessary information using a mobile device without going through store staff, thereby improving the customer experience.

[0702] "Industry-specific information" refers to information related to a particular industry, and is knowledge and data that is only useful within that industry.

[0703] "Preprocessing" refers to the initial data processing steps performed to prepare raw data into a format that can be easily handled by machine learning algorithms.

[0704] A "machine learning algorithm" refers to a technique that uses data to train a model and then makes predictions or classifications based on new data.

[0705] A "pre-trained algorithm" is a machine learning model that has already been trained and is ready to handle a specific task.

[0706] An "inquiry" refers to a question or request made by a user to obtain information.

[0707] The "Answer Information Repository" is a database that stores past inquiries and their corresponding answers.

[0708] A "generative AI algorithm" is an artificial intelligence technology that generates natural language responses based on input data.

[0709] "Mobile devices" refer to portable computer devices such as smartphones and tablets.

[0710] "In-store information" refers to business information related to a specific physical store, such as products, inventory, and promotions.

[0711] The system for implementing this invention consists of a server, a customer's mobile terminal, and a store information management system. The server collects industry-specific information through online databases and information provision interfaces and preprocesses it. This prepares the data in a format that can be efficiently handled by machine learning algorithms. The preprocessed data is then trained by the server using machine learning algorithms and used to automatically generate inquiries and their answers.

[0712] The trained algorithm uses natural language processing techniques to accurately understand industry-specific terminology and concepts and generate responses. When a customer makes a query using a mobile device, the server refers to the relevant query and answer database to provide a corresponding response. If a matching answer does not exist in the database, the generation AI algorithm generates a supplementary answer in real time and displays it on the customer's device.

[0713] For example, if a customer in a store enters "Please tell me about the current special sale in the store" into a mobile terminal, the server retrieves the latest sale information from its database and provides an accurate response based on that information.

[0714] An example of a prompt for the generative AI model in this case is as follows:

[0715] "Could you tell me about any sales currently running in your store? I'd like to know more about the latest discounts and campaigns."

[0716] This allows customers to quickly and accurately obtain the information they need without having to ask store staff directly.

[0717] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0718] Step 1:

[0719] The server collects industry-specific information through online databases and information provision interfaces. It retrieves information from specified APIs as input and converts unstructured data into an easily understandable format. The output is data that has been prepared for preprocessing.

[0720] Step 2:

[0721] The server preprocesses the collected data. The input is the data prepared in step 1. Data cleaning, tokenization, and feature extraction are performed to transform the data into a format usable by machine learning algorithms. Noise reduction and filtering of necessary data are also performed during this process. The output becomes the training dataset.

[0722] Step 3:

[0723] The server uses pre-processed data to train a machine learning algorithm. It uses a training dataset as input and learns patterns using algorithms such as neural networks. The output is a trained algorithm that can be used to generate queries and answers.

[0724] Step 4:

[0725] A user enters a specific query from a mobile device. The input is natural language text received through the device's interface. The device sends the entered text to the server. The output is the request to the server.

[0726] Step 5:

[0727] The server receives a query from a user and consults its answer database. The input is text sent by the user. The server uses natural language processing techniques to search the database for existing answers related to the query. The output is the answer as a search result.

[0728] Step 6:

[0729] If the answer does not exist in the answer database, the server uses a generative AI model to complete the answer. The input is the query for which no match was found in step 5. The generative AI model uses a prompt to generate a response in natural language. The output is the generated answer text.

[0730] Step 7:

[0731] The server returns the results to the user's terminal. The input is the answer obtained in step 5 or step 6. The terminal displays the received answer on the screen and informs the user. The output is the information displayed to the user.

[0732] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0733] This invention realizes a system that provides more personalized answers by combining an AI-based FAQ system using industry-specific data with a function that recognizes user emotions. This system comprises a server, terminals, and an emotion engine, and provides industry-savvy answers instantly while adjusting the tone and content according to the user's emotional state.

[0734] Data collection and model learning

[0735] The server collects industry-specific data through online databases and APIs. The collected data is preprocessed, and industry expertise is learned using machine learning models. These models utilize natural language processing techniques to understand technical terms and complex concepts.

[0736] FAQ generation and management

[0737] The server uses a pre-trained model to automatically generate frequently asked questions and their answers. The generated FAQs are stored in a database and regularly updated based on the latest information.

[0738] Emotion recognition and response

[0739] When a user enters a question via their device, the emotion engine analyzes the user's wording and, if possible, their tone of voice to identify their emotions. For example, if the user is feeling anxious, the engine adjusts the response to alleviate those emotions.

[0740] User response

[0741] The server selects the most relevant answer from the FAQ database, but adjusts the tone of the answer based on insights provided by the sentiment engine. For example, if the user is stressed, the server will provide an answer using gentler language.

[0742] Specific example

[0743] If a user types "What are some ways to deal with the recent rise in interest rates?" on their device and senses anxiety, the server uses an emotion engine to identify the anxiety and provides a reassuring response such as, "If you have a detailed plan regarding interest rate countermeasures, please consult your assigned advisor. They can handle it calmly."

[0744] In this way, communication that takes user emotions into account becomes possible, resulting in a more user-friendly system.

[0745] The following describes the processing flow.

[0746] Step 1:

[0747] The server collects industry-specific data from online databases and APIs. This data includes industry news, regulatory information, and technical documents.

[0748] Step 2:

[0749] The server preprocesses the collected data, removing duplicates, denoising noise, and standardizing the format as needed.

[0750] Step 3:

[0751] The server passes pre-processed data to a machine learning model, which then learns industry-specific knowledge. This model uses natural language processing to understand technical terms and context.

[0752] Step 4:

[0753] The server automatically generates frequently asked questions and their answers using a pre-trained model and stores them in an FAQ database.

[0754] Step 5:

[0755] The user enters a question via their device. For example, they might enter a specific question in natural language, such as, "Tell me about the latest security measures."

[0756] Step 6:

[0757] The terminal sends the entered question to the server. The server searches the FAQ database and identifies the relevant answer.

[0758] Step 7:

[0759] The emotion engine determines the user's emotions from the user's input text and voice. Here, it identifies emotional states such as anxiety, excitement, and anger.

[0760] Step 8:

[0761] The server adjusts the tone and content of the response based on the results of the emotion engine. For example, if the user is nervous, it will generate a response in a calm tone.

[0762] Step 9:

[0763] The server sends the adjusted response to the terminal.

[0764] Step 10:

[0765] The terminal displays the response received from the server to the user. The user can then ask further questions based on this information.

[0766] This processing flow allows users to quickly receive professional and emotionally sensitive responses.

[0767] (Example 2)

[0768] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0769] While conventional AI-based FAQ systems can provide industry-specific information quickly and accurately, they lack the ability to provide personalized responses that consider user emotions. It is necessary to adjust the tone of responses according to the user's feelings to achieve communication that not only provides information but also considers the user's feelings.

[0770] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0771] In this invention, the server includes means for collecting industry-specific information, means for preprocessing the information and training it using a machine learning algorithm, and means for analyzing the user's emotional state and adjusting the representation of the solution according to the user's emotions. This enables not only the provision of information but also a personalized user experience that takes emotions into consideration.

[0772] "Industry-specific information" refers to specialized knowledge and data used within a particular industry, and is generally not widely used in other industries.

[0773] "Preprocessing" is the process of converting raw data into a format that can be used by algorithms, and includes tasks such as denoising, normalization, and data cleaning.

[0774] A "machine learning algorithm" is a computational method that learns patterns from past data and makes predictions and decisions based on new data.

[0775] A "trained algorithm" refers to a machine learning algorithm that has been trained using a specific dataset and has acquired a certain degree of predictive and decision-making ability.

[0776] A "generative AI algorithm" refers to artificial intelligence technology that automatically creates new data or answers based on human instructions.

[0777] "Natural language processing technology" is a technology that enables computers to understand, generate, and translate human language.

[0778] "User emotional state" refers to the psychological state a user is in when seeking information, and includes emotions such as anxiety, excitement, and tension.

[0779] This invention relates to a system that provides personalized responses that take user emotions into account, using an AI system based on industry-specific information. The main components of this system include a server, a terminal, and an emotion recognition engine.

[0780] The server plays a central role in information gathering, model training, and response generation. Specifically, the server collects industry-specific information using remote databases and application programming interfaces (APIs). The collected information is then cleaned and filtered, and used to train a model with machine learning algorithms. This model utilizes natural language processing techniques (such as BERT and GPT) to understand industry jargon and context.

[0781] When a user enters a question via their device, the information is sent to a server, which then searches the FAQ database for relevant answers. During this process, a generative AI algorithm is used to generate new answers even for questions not found in the database.

[0782] The emotion recognition engine analyzes the user's text and voice tone to identify their emotional state. The server uses this insight to adjust the tone of its response before sending it back to the device. For example, if a user makes an anxious input such as, "What should I do about the recent rise in interest rates?", the system will provide a reassuring response such as, "If you have a detailed plan for dealing with interest rates, please consult your advisor. They can handle it calmly."

[0783] As an example of a prompt, one could instruct the AI ​​generation algorithm to "compile a list of frequently asked questions about the industry." This would enable the provision of information that takes into account the user's emotional aspects, resulting in a more user-friendly experience.

[0784] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0785] Step 1:

[0786] The server collects industry-specific information. Specifically, the server retrieves necessary data using remote databases and APIs. The input is information from online sources, and the output is structured data stored on the server's storage. After data retrieval, the server performs data preprocessing such as filtering duplicate data and formatting.

[0787] Step 2:

[0788] The server collects and preprocesses data to train machine learning algorithms. The input is preprocessed industry data, and the output is a trained model capable of understanding industry-specific challenges. This process utilizes natural language processing techniques to improve terminology understanding and context recognition performance.

[0789] Step 3:

[0790] The server uses a pre-trained model to automatically generate frequently asked questions and their answers. It takes the trained model and the prompt "Please compile a list of frequently asked questions about the industry" as input, and the output is a collection of answers stored in an FAQ database. During this process, quality checks are performed to ensure that the automatically generated answers meet specific criteria.

[0791] Step 4:

[0792] The user enters a question into the system via a terminal. The input is text data received from the terminal, and this data is sent to the server. The output is the user's inquiry sent to the server.

[0793] Step 5:

[0794] The server consults an FAQ database and selects the most relevant answer to the user's question. The input is the user's question and the FAQ database, and the output is the most relevant answer. In this process, a generative AI model is used to supplement and generate new answers for questions not found in the database.

[0795] Step 6:

[0796] The emotion recognition engine analyzes the user's emotional state from their input. The input is either text or voice data from the user, and the output is identified emotional information. Based on this information, the server adjusts the tone of its response.

[0797] Step 7:

[0798] The server sends an emotionally balanced response to the device and presents it to the user. The input is the balanced response, and the output is the response displayed to the user on the device. This allows the user to receive information that takes emotional aspects into consideration.

[0799] (Application Example 2)

[0800] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0801] Electronic payment services require prompt and appropriate responses to user concerns and questions during transactions. However, conventional FAQ systems fail to consider the user's emotional state and can only provide standard answers, resulting in insufficient resolution of user anxieties. Furthermore, providing answers that take emotions into account can improve the user experience.

[0802] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0803] In this invention, the server includes means for collecting industry-specific data, means for training a machine learning model, means for automatically generating frequently asked questions and their answers, means for recognizing the user's emotional state and adjusting the tone of the answers accordingly, and means for application in an electronic payment service. As a result, it becomes possible to provide personalized answers that correspond to the user's emotional state, thereby reducing user anxiety and improving their sense of security.

[0804] "Industry-specific data" refers to a collection of information and knowledge related to a particular industry or sector.

[0805] A "machine learning model" is an algorithm that automatically learns patterns and knowledge from data.

[0806] "Frequently asked questions" are common questions that users repeatedly ask.

[0807] "Automatic generation" refers to the process by which a computer creates questions and answers without human intervention.

[0808] "User emotional state" refers to the psychological reactions and feelings of a user at a specific point in time.

[0809] "The tone of the response" refers to the style of language and expression used in the response provided.

[0810] An "electronic payment service" is a digital payment system for transactions conducted over the internet.

[0811] "Personalized responses tailored to emotional state" refer to individually optimized responses that are adjusted based on the user's emotions.

[0812] This invention is for implementing an FAQ system that responds to user emotions in electronic payment services. The main components of the system include a server, a terminal, and an emotion engine.

[0813] The server first collects industry-specific data from online sources. The collected data is preprocessed and trained on a machine learning model. This model uses platforms such as TensorFlow to deepen its understanding of industry-specific terminology and concepts and automatically generate answers to frequently asked questions.

[0814] When a user enters a question via their device, that information is sent to the server. The emotion engine then analyzes the user's input and uses the Google Cloud Natural Language API to identify their emotional state. For example, if it detects that the user is feeling anxious, it adjusts the tone of the response generated by the server based on that result.

[0815] Ultimately, the server retrieves the most relevant answers from the FAQ database and provides them in language appropriate to the user's emotional state. This allows the user to have a more reassuring experience.

[0816] For example, when a user types "My transfer hasn't been reflected. What should I do?", the server, after having its emotion engine identify the user's anxiety, provides a personalized response in a friendly tone, such as "We understand your concern. It usually takes a few hours, but if it's still not reflected, please contact customer support."

[0817] An example of a prompt for the generating AI model is: "Generate a response to a user who is anxious because their payment has not been reflected, using an appropriate tone based on the user's emotions."

[0818] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0819] Step 1:

[0820] The terminal receives a question from the user. The user enters the question through the terminal's interface, and this input data is sent to the server. The input data is in text format and needs to be properly transmitted over the network for further processing.

[0821] Step 2:

[0822] The server analyzes the user input it receives. The text of the input data is sent to the sentiment engine, which performs sentiment analysis using the Google Cloud Natural Language API. Here, natural language processing techniques are used to identify the user's emotional state (e.g., anxiety, relief, tension) from the input text. The output is data indicating the emotional state.

[0823] Step 3:

[0824] The server searches the FAQ database based on the sentiment analysis results. The server operates a machine learning model (using TensorFlow) to automatically search and extract relevant questions and answers from the FAQ database. In this step, the user's question and analyzed sentiment state are taken as input, and the most appropriate answer candidates are obtained as output.

[0825] Step 4:

[0826] The server adjusts the tone of the response. Using an AI model that generates responses, and referencing the emotional state obtained by the emotion engine, the response tone is harmoniously and appropriately adjusted. Specifically, if the emotion is anxiety, reassuring words and expressions are selected. Here, the input is the emotional state and response data from FAQs, and the output is the final adjusted response.

[0827] Step 5:

[0828] The terminal presents the user with the adjusted response received from the server. The final response is displayed to the user through the terminal's interface, completing the handling of the user's inquiry. This includes data transmission from the server to the terminal, and appropriate feedback is provided to the user as a result of the processing.

[0829] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0830] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0831] 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 this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0832] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0833] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0834] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0835] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0836] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0837] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0838] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0839] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0840] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0841] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0842] 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.

[0843] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0844] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0845] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0846] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0847] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0848] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0849] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0850] The following is further disclosed regarding the embodiments described above.

[0851] (Claim 1)

[0852] Means of collecting industry-specific data,

[0853] A means for preprocessing the aforementioned data and training a machine learning model,

[0854] A means for automatically generating frequently asked questions and their answers using the aforementioned trained model,

[0855] A means for receiving questions from users and providing relevant answers by referring to the generated question and answer database,

[0856] The system includes means for supplementing answers to questions not present in the aforementioned database using a generative AI model.

[0857] (Claim 2)

[0858] The system according to claim 1, wherein the data collection means is configured to acquire information through an online database or API.

[0859] (Claim 3)

[0860] The system according to claim 1, wherein the trained model is configured to understand industry-specific terminology and concepts using natural language processing technology and generate answers.

[0861] "Example 1"

[0862] (Claim 1)

[0863] Information aggregation means,

[0864] A means for processing the aforementioned information and training it using a machine learning algorithm,

[0865] A means for automatically generating frequently asked questions and their answers using the aforementioned trained algorithm,

[0866] A means for receiving inquiries from users and providing relevant answers by referring to the generated question and answer information sources,

[0867] For inquiries not found in the aforementioned information sources, a means of supplementing the answer using a generative AI algorithm is provided.

[0868] A system that includes means for rapidly distributing information via a user interface when providing the aforementioned answer.

[0869] (Claim 2)

[0870] The system according to claim 1, wherein the information aggregation means is configured to acquire information through a network database or a program interface.

[0871] (Claim 3)

[0872] The system according to claim 1, wherein the pre-trained algorithm uses natural language processing technology to understand industry-specific terminology and concepts and generate answers.

[0873] "Application Example 1"

[0874] (Claim 1)

[0875] Means of collecting industry-specific information,

[0876] A means for preprocessing the aforementioned information and training it using a machine learning algorithm,

[0877] A means for automatically generating frequently asked questions and their answers using the aforementioned trained algorithm,

[0878] A means for receiving inquiries from users and providing relevant responses by referring to the generated inquiry and answer information database,

[0879] For inquiries not present in the aforementioned information repository, a means is provided to supplement the response using a generation AI algorithm,

[0880] A system that allows a customer to inquire about in-store information from a mobile terminal and includes means for displaying the response on the mobile terminal.

[0881] (Claim 2)

[0882] The system according to claim 1, wherein the information gathering means is configured to acquire information through an electronic database or an information provision interface.

[0883] (Claim 3)

[0884] The system according to claim 1, wherein the trained algorithm is configured to understand industry-specific terminology and concepts using language processing techniques and generate a response.

[0885] "Example 2 of combining an emotion engine"

[0886] (Claim 1)

[0887] Means of collecting industry-specific information,

[0888] A means for preprocessing the aforementioned information and training it using a machine learning algorithm,

[0889] A means for automatically generating frequently asked questions and their solutions using the aforementioned trained algorithm,

[0890] A means for receiving inquiries from users and providing relevant solutions by referring to the generated database of issues and solutions,

[0891] For problems not present in the aforementioned database, a means of supplementing solutions using a generative AI algorithm is employed.

[0892] A means for analyzing the user's emotional state and adjusting the expression of the aforementioned solution according to the user's emotions,

[0893] A system that includes this.

[0894] (Claim 2)

[0895] The system according to claim 1, wherein the information gathering means is configured to acquire data through a remote database or an application programming interface.

[0896] (Claim 3)

[0897] The system according to claim 1, wherein the trained algorithm is configured to use natural language processing techniques to understand industry-specific terminology and concepts and generate solutions.

[0898] "Application example 2 when combining with an emotional engine"

[0899] (Claim 1)

[0900] Means of collecting industry-specific data,

[0901] A means for preprocessing the aforementioned data and training a machine learning model,

[0902] A means for automatically generating frequently asked questions and their answers using the aforementioned trained model,

[0903] A means for receiving questions from users and providing relevant answers by referring to the generated question and answer database,

[0904] For questions not present in the aforementioned database, a means of supplementing the answer using a generative AI model is provided.

[0905] A means of recognizing the user's emotional state and adjusting the tone of the response based on that,

[0906] A system applied to electronic payment services that includes means to enhance user confidence.

[0907] (Claim 2)

[0908] The system according to claim 1, wherein the data collection means is configured to acquire information through an online information source.

[0909] (Claim 3)

[0910] The system according to claim 1, wherein the trained model uses language processing technology to understand industry-specific terminology and concepts and generate answers. [Explanation of Symbols]

[0911] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Means of collecting industry-specific data, A means for preprocessing the aforementioned data and training a machine learning model, A means for automatically generating frequently asked questions and their answers using the aforementioned trained model, A means for receiving questions from users and providing relevant answers by referring to the generated question and answer database, The system includes means for supplementing answers to questions not present in the aforementioned database using a generative AI model.

2. The system according to claim 1, wherein the data collection means is configured to acquire information through an online database or API.

3. The system according to claim 1, wherein the trained model is configured to understand industry-specific terminology and concepts using natural language processing technology and generate answers.

Images