System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting help information of a plurality of online platforms; means for analyzing the collected information and classifying the collected information into a specific category; means for training a natural language processing model based on the classified information; means for analyzing a user question using natural language analysis; means for generating an appropriate answer to the analyzed question; and means for providing the generated answer to a user terminal.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] With the spread of the internet, opportunities for individuals to sell goods using online platforms are increasing. However, because each online platform has different sales procedures, fees, delivery methods, and other information, individuals who use multiple platforms require time and knowledge to adapt to each platform. Furthermore, each platform's customer support can only respond to questions about its own services, making it difficult for users to obtain cross-sectional information about each platform. Obtaining such information is a major barrier, especially for first-time users, making it difficult to conduct sales activities smoothly. [Means for solving the problem]

[0005] The present invention provides a system that collects help information from multiple online platforms, analyzes and classifies it using natural language processing, and provides appropriate answers to user questions. Specifically, the system collects official help information from multiple online platforms via scraping or API, analyzes the information, and classifies it into specific categories (e.g., listing methods, fees, packaging methods, etc.). A natural language processing model is then trained based on this classification information. When a user inputs a question, the system analyzes the question using natural language processing, searches a database for relevant information based on the analysis results, and generates and provides an appropriate answer to the user. This system allows users to easily obtain cross-sectional information and comparative information about each platform, facilitating smooth sales activities. Specifically, it is possible to provide comparative information on fees for multiple platforms and specific instructions on packaging methods.

[0006] "Online platform" is a collective term for websites and applications that support e-commerce and are designed to sell goods and services over the Internet.

[0007] "Help information" refers to support materials such as instructions, FAQs, and usage guidelines that users refer to when using online platforms.

[0008] "Scraping" is a technique that uses a program to automatically retrieve website content and store it in a database.

[0009] "API" stands for Application Programming Interface, an interface for sharing data and functions between different software systems.

[0010] "Natural language processing" is a technology that uses computers to analyze, understand, and generate natural language used by humans.

[0011] "Analysis" is the process of examining and investigating collected data and information in detail to understand its contents.

[0012] "Classification" is the process of organizing and classifying analyzed information according to specific criteria or categories.

[0013] "Learning" is the process by which an AI model understands certain patterns and rules based on given data, enabling it to make future predictions and decisions.

[0014] A database is a system for systematically storing and managing data in digital form, allowing you to quickly search and obtain the information you need.

[0015] "Question analysis" is the process of understanding the intent and purpose of a question entered by a user and generating an appropriate response based on that content.

[0016] "Answer generation" is the process of constructing appropriate and relevant information to the user's question based on the analysis results.

[0017] A "chatbot" is a software application designed to interact with users in a natural, conversational manner.

[0018] "User Device" means a device used by a User to access the System and input or receive information, including a computer, smartphone, tablet, etc. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3]FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0020] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0021] First, the terms used in the following description will be explained.

[0022] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0023] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0024] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0025] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0026] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0031] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0032] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0033] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0038] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0039] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0040] The present invention provides a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms.

[0041] System Overview

[0042] The system is mainly composed of the following three roles (server, terminal, and user).

[0043] 1. The server collects information from the online platform, analyzes it, stores it in a database, receives questions from users, generates answers to those questions, and sends them to the device.

[0044] 2. The terminal provides an interface for the user to enter questions and displays the answers received from the server.

[0045] 3. The user accesses the system through a terminal, enters a question, and obtains the required information.

[0046] Program processing

[0047] 1. Data Collection

[0048] The server collects information from the official help pages of multiple online platforms, using scraping technology and APIs.

[0049] 2. Data Analysis and Classification

[0050] The server analyzes the collected help information using natural language processing (NLP) and classifies each piece of information into specific categories (e.g., listing methods, fees, packaging methods, etc.).

[0051] 3. Training the AI ​​model

[0052] The server uses the classified information to train an AI model, which improves its ability to generate appropriate answers to user questions.

[0053] 4. Receiving and analyzing questions

[0054] When a user uses a device to enter a question, the question is sent to a server, which receives the question and analyzes its intent using natural language processing.

[0055] 5. Answer Generation

[0056] The server searches for relevant information from a database based on the parsed question and generates an appropriate answer.

[0057] 6. Providing answers

[0058] The server sends the generated answer to the terminal, which displays it to the user.

[0059] Specific examples

[0060] Example 1: When a user asks, "What are the fees for Platform A and Platform B?"

[0061] Data collection

[0062] The server collects fee information from the official help pages of Platform A and Platform B.

[0063] Data Analysis and Classification

[0064] The server analyzes the collected fee information and categorizes it for each platform.

[0065] Receiving and parsing questions

[0066] When a user types "What are the fees for Platform A and Platform B?" into a terminal, the question is sent to the server, which analyzes the question and determines that the user is seeking comparative information on fees.

[0067] Generate answers

[0068] The server searches the database for commission information for Platform A and Platform B and generates a response such as, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[0069] Providing answers

[0070] The generated answers are sent to the device and displayed to the user, who can use them to decide which platform to use.

[0071] Example 2: When a user asks, "How do I pack on platform C?"

[0072] Data collection

[0073] The server collects information about packaging methods from the official help page of platform C.

[0074] Data Analysis and Classification

[0075] The server analyzes the collected packaging method information and classifies it into specific steps and recommended packaging materials.

[0076] Receiving and parsing questions

[0077] When a user types "Tell me how to pack on platform C" into a terminal, the question is sent to the server, which analyzes the question and determines that it is asking for information about packing methods.

[0078] Generate answers

[0079] The server searches a database for information about platform C's packaging method and generates a specific answer such as "Platform C's packaging method is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape."

[0080] Providing answers

[0081] The generated answer is sent to the device and displayed to the user, who can use it to properly pack the items.

[0082] As described above, by using the system of the present invention, users can easily obtain cross-sectional information on multiple online platforms, which allows individual sellers to sell more smoothly and eliminates literacy gaps when using various platforms.

[0083] The processing flow will be explained below.

[0084] Step 1:

[0085] Finalize the help page

[0086] The server lists the official help page URLs of multiple online platforms.

[0087] Specifically, links to help pages are extracted from the official websites of each platform, and the URLs are registered in a database.

[0088] Step 2:

[0089] Information scraping

[0090] The server accesses the listed URLs, performs scraping, and obtains text information from the pages.

[0091] Specifically, it retrieves HTML content from a specified URL and analyzes and extracts text data from it.

[0092] Step 3:

[0093] Storage in the database

[0094] The server stores the acquired information in a database as structured data, organizing the data into categories that differ for each reuse service (e.g., listing method, handling fee, packaging method, etc.).

[0095] Specifically, through analysis, each item is registered in a database as an independent record.

[0096] Step 4:

[0097] Analysis using natural language processing

[0098] The server analyzes the stored text data using natural language processing (NLP) algorithms to extract key keywords and phrases.

[0099] Specifically, it performs processes such as word segmentation, part-of-speech tagging, and entity recognition to structure the information.

[0100] Step 5:

[0101] Information extraction and classification

[0102] Based on the analysis results, the server classifies and tags information such as the listing process, fees, and packaging methods for each reuse service.

[0103] Specifically, the information is assigned to an appropriate category based on the extracted keywords and phrases.

[0104] Step 6:

[0105] Training an AI model

[0106] The server uses the classified information to train the chatbot's AI model.

[0107] Specifically, information from the database is used to train a model using a machine learning algorithm (e.g., deep learning).

[0108] Step 7:

[0109] Chatbot display

[0110] The device displays the chatbot's user interface, which can be provided in a web browser, a mobile app, or a desktop application.

[0111] Specifically, when a user accesses the site, it runs a program that pops up a chat window.

[0112] Step 8:

[0113] User question input

[0114] Users enter questions related to the listing process into the chatbot's input field.

[0115] Specifically, the user enters a question using a keyboard or voice input and sends it through the chatbot interface.

[0116] Step 9:

[0117] Receiving questions

[0118] The server receives the question entered by the user.

[0119] Specifically, it receives data sent via the chat window in real time and begins the analysis process.

[0120] Step 10:

[0121] Intent Analysis

[0122] The server analyzes the received question using a natural language processing algorithm to determine the user's intent.

[0123] Specifically, a question-and-answer model is used to identify the type of information being sought from the question.

[0124] Step 11:

[0125] Search for related information

[0126] The server searches the database for relevant information based on the analysis results.

[0127] Specifically, the identified categories and keywords are used to quickly search for corresponding data in the database.

[0128] Step 12:

[0129] Answer structure

[0130] The server uses the search results to construct a specific answer to the user's question. If comparative information is required, the server combines data on multiple reuse services to create the answer.

[0131] Specifically, data is embedded in a response template to generate an easy-to-understand and well-organized response.

[0132] Step 13:

[0133] Submit your answer

[0134] The server then sends the constructed response to the user's terminal.

[0135] Specifically, the generated answer message is sent to the user via the chatbot interface.

[0136] Step 14:

[0137] Show Answers

[0138] The terminal displays the received response on the chatbot's interface.

[0139] Specifically, received messages will be displayed in the chat window so that users can view them immediately.

[0140] The above are the detailed processing steps of the system. This system allows users to efficiently obtain information and conduct sales activities utilizing multiple reuse services.

[0141] Example 1

[0142] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0143] In the past, collecting and analyzing auxiliary information from multiple electronic platforms and providing quick and accurate answers to user questions required a lot of time and effort. In particular, advanced technology was required to efficiently collect and properly analyze information from each platform. Furthermore, when a user's question related to a specific platform, it was difficult to accurately generate an answer. Therefore, the present invention aims to solve these problems and enable users to easily obtain information from various platforms.

[0144] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0145] In this invention, the server includes a means for collecting auxiliary information from multiple electronic platforms, a means for analyzing the collected information and classifying it into specific categories, and a means for training a natural language processing algorithm based on the classified information, thereby enabling the server to provide quick and accurate answers to questions entered by users.

[0146] An "electronic platform" is a general term for services and systems provided online, and refers to the infrastructure through which users can obtain information and conduct transactions.

[0147] "Supporting information" refers to all information such as guidelines, FAQs, manuals, etc. that electronic platforms provide to users.

[0148] "Means of collection" refers to the technologies and methods used to obtain auxiliary information from electronic platforms, including the use of scraping technologies and APIs.

[0149] "Means for analyzing and classifying into specific categories" refers to the method of analyzing the collected auxiliary information using data processing techniques and organizing it into relevant categories.

[0150] "Natural language processing algorithms" refers to a set of computational techniques that enable computers to understand and process human language, including, for example, machine learning models and dictionary-based methods for analyzing the meaning of sentences.

[0151] "Training" refers to the way algorithms are used to train AI models based on classified information, allowing the system to generate more accurate answers.

[0152] "Means for analyzing user input" refers to technology that uses natural language processing technology to analyze questions or requests entered by users and understand their intent and content.

[0153] "Answer generation means" refers to techniques or methods for generating appropriate answers based on analyzed user input, including database searches and the use of generative AI models.

[0154] "Means of providing" refers to the interface or communication means for displaying or communicating the generated answer to the user.

[0155] "Scraping technology" refers to the technique of analyzing the HTML structure of a website and extracting specific information programmatically.

[0156] "API" stands for Application Programming Interface, and refers to an interface that allows software to communicate with each other.

[0157] A "library" refers to a collection of code and data for shared use of program functions and data, and natural language processing libraries include existing analysis algorithms and datasets.

[0158] "JSON format" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data exchange format for defining data structures.

[0159] This system collects, analyzes, and classifies auxiliary information from multiple electronic platforms, enabling it to respond quickly and accurately to user inquiries. The system is primarily comprised of three roles: a server, a terminal, and a user.

[0160] server

[0161] The server mainly performs the following tasks:

[0162] 1. Data Collection

[0163] The server uses scraping techniques and APIs to collect auxiliary information from electronic platforms. Specifically, it uses Python's BeautifulSoup library to analyze the HTML structure of web pages and extract the necessary information. It can also use appropriate platform APIs to obtain official information.

[0164] 2. Data Analysis and Classification

[0165] The server analyzes the collected auxiliary information using natural language processing (NLP) techniques, such as SpaCy and NLTK, to analyze the meaning of the text data and organize the information into specific categories (e.g., listing method, handling fee, packaging method).

[0166] 3. Training the AI ​​model

[0167] The server trains an AI model based on the classified information. Machine learning libraries such as Scikit-learn and TensorFlow are used to build the AI ​​model. The training dataset consists of the collected and classified auxiliary information. The trained AI model is then capable of generating appropriate answers to user questions.

[0168] 4. Receiving and analyzing questions

[0169] When a user uses a device to input a question, the question is sent to the server, which uses natural language processing (NLP) to analyze the received question and determine its intent. For example, if a user inputs "What are the fees for platform A and platform B," the server understands that the user is seeking comparative information on fees.

[0170] 5. Answer Generation

[0171] The server searches for relevant information from a database based on the parsed question and generates an appropriate answer for the user, which is then sent to the device in JSON format.

[0172] Terminal

[0173] The terminal provides an interface for users to enter questions and displays the answers received from the server. Specifically, when a user enters a question in a text input field and presses the submit button, the question is sent to the server as an HTTP request. After receiving the answer from the server, it is displayed in an easy-to-understand manner to the user using HTML and CSS.

[0174] user

[0175] Users access the system through their terminals, input questions, and obtain the necessary information. For example, if a user asks, "Tell me how to pack on platform C," the question is analyzed by the server, and relevant information is generated and displayed on the terminal.

[0176] Specific examples

[0177] For example, if a user asks "What are the fees for Platform A and Platform B?", the following will show:

[0178] The user enters a question into the text input field on the device and presses the send button.

[0179] The device sends the question to the server as an HTTP request.

[0180] The server receives the question and analyzes it using natural language processing.

[0181] The server searches the database for the commission information of Platform A and Platform B and generates a response saying, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[0182] The server sends the generated response in JSON format to the terminal.

[0183] The device displays the received response to the user.

[0184] As described above, by using the system of the present invention, users can easily obtain cross-sectional information on multiple electronic platforms, which allows individual sellers to conduct sales activities more smoothly and eliminates literacy differences when using various platforms.

[0185] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0186] Step 1: Data collection

[0187] Input: URLs or API endpoints for multiple Electron platforms

[0188] Specific operation: The server uses Python's BeautifulSoup to retrieve the HTML of each platform's official help page and extract the necessary information. It also retrieves information using the platform's API.

[0189] Data processing: Extract appropriate text data from the acquired HTML or API response.

[0190] Output: Text data of extracted auxiliary information

[0191] Step 2: Data analysis and classification

[0192] Input: Text data obtained in the data collection step

[0193] What it does: The server uses a natural language processing (NLP) library (e.g., SpaCy, NLTK) to parse the text data, extracting keywords from the text and organizing each piece of information into specific categories (e.g., listing method, handling fee, packaging method).

[0194] Data processing: Based on the analysis, the information is classified into categories.

[0195] Output: A set of classified auxiliary information

[0196] Step 3: Training the AI ​​model

[0197] Input: A set of classified auxiliary information

[0198] Specific operation: The server trains an AI model using a machine learning library (e.g., Scikit-learn, TensorFlow). It uses the classification data as training data and learns input and output patterns.

[0199] Data processing: Extract features from input data and train the model.

[0200] Output: Trained AI model

[0201] Step 4: Receiving the question

[0202] Input: User question (text format)

[0203] Specific operation: The user enters a question into the device interface and presses the send button. The device then sends the question to the server as an HTTP request.

[0204] Data processing: None

[0205] Output: The question sent to the server

[0206] Step 5: Parsing the Question

[0207] Input: The question sent to the server

[0208] How it works: The server uses natural language processing (NLP) techniques to analyze the question and determine its intent. For example, if a user types, "What are the fees for platform A and platform B?", the server understands that the user is looking for comparative information on fees.

[0209] Data processing: Based on the analysis results, identify the intent of the question.

[0210] Output: Parsed question intent

[0211] Step 6: Generate an answer

[0212] Input: Parsed question intent

[0213] Specific operation: The server searches the database for relevant information and generates an appropriate answer. For example, it searches the database for information on "commission fees" and generates an answer such as "Platform A's commission fee is 10% of the sales price, and Platform B's commission fee is 12% of the sales price."

[0214] Data processing: Extracting relevant information and structuring answers.

[0215] Output: Generated answer (text format)

[0216] Step 7: Provide your answers

[0217] Input: Generated answer (text format)

[0218] Specific operation: The server generates a response and sends it to the device in JSON format. The device receives the response and displays it in an easy-to-understand manner for the user using HTML and CSS.

[0219] Data processing: Convert JSON format data into HTML format so that it can be displayed.

[0220] Output: The answer shown to the user

[0221] (Application example 1)

[0222] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0223] When using help information on multiple online platforms, users often find it difficult to quickly and accurately obtain the information they need. This is especially true for users without specialized knowledge, as each platform has different specifications and usage methods. Furthermore, existing systems lack the analytical capabilities to match questions with appropriate answers, potentially resulting in reduced user satisfaction. To address these challenges, a system is needed that allows users to efficiently obtain information and provide quick and accurate answers to their questions.

[0224] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0225] In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a natural language processing model based on the classified information, means for sending questions to the system via prompt sentences, and means for searching the collected help information on the server and generating optimal answers for the user terminal, thereby enabling the server to generate quick and accurate answers to user questions using help information from multiple online platforms.

[0226] An "online platform" is a system that provides services and content via the Internet.

[0227] "Help Information" refers to documentation and FAQs to help users better understand how to use and troubleshoot the Platform.

[0228] "Means of collection" refers to the function of obtaining and organizing information from the Internet.

[0229] "Means of analysis" refers to the process of converting collected information into an understandable format.

[0230] "Categorizing" means separating information into themes.

[0231] A "natural language processing model" is a machine learning algorithm for analyzing text data and understanding its meaning.

[0232] A "prompt" is text that the user enters into their system to ask a question or request.

[0233] The "means for sending a question" is a function for sending the user's input to the server.

[0234] The "means for generating the optimal answer" is a process for generating the most appropriate answer to the user's question.

[0235] A "user terminal" is a device through which a user can enter questions and receive answers.

[0236] This invention provides a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms. The system consists of three main components: a server, a terminal, and a user.

[0237] Server Configuration

[0238] 1. Data Collection

[0239] The server collects information from the official help pages of multiple online platforms using scraping technology and APIs.

[0240] 2. Data Analysis and Classification

[0241] The server analyzes the collected help information using natural language processing (NLP) and classifies it into specific categories (e.g., account management, payment issues, content management, etc.).

[0242] 3. Training the AI ​​model

[0243] The classified information is used to train an AI model (e.g., BERT, GPT, etc.), which improves its ability to generate appropriate answers to user questions.

[0244] 4. Receiving the prompt

[0245] Prompts are used to allow users to send questions to the system using their devices, such as "How do I delete my account?" or "What are the benefits of a premium membership?"

[0246] 5. Answer Generation

[0247] After receiving the prompt, the server generates the best answer based on the collected help information, and AI models are heavily utilized in this process.

[0248] 6. Providing answers

[0249] The generated answer is sent to the user's terminal and displayed to the user.

[0250] Device configuration

[0251] 1. User Interface

[0252] It provides a chat-style interface for users to enter questions.

[0253] 2. Data Transmission

[0254] The terminal sends the user's question to the server as a prompt sentence.

[0255] 3. Answer display

[0256] Receives the response sent by the server and displays it to the user.

[0257] User Actions

[0258] 1. Enter your question

[0259] The user uses the terminal to enter a question for help information as a prompt sentence.

[0260] 2. Receiving and Confirming Responses

[0261] The user checks the answer sent by the server and obtains the necessary information. For example, if the user enters the question "How do I delete my account?", the server generates a specific answer such as "Go to the settings screen and select the account deletion option."

[0262] This system allows users to quickly and accurately obtain information about multiple online platforms, significantly reducing the effort required for users to efficiently understand and use different platforms.

[0263] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0264] Step 1: Data collection

[0265] The server collects information from the official help pages of multiple online platforms using scraping techniques (e.g., BeautifulSoup, Scrapy) and APIs (e.g., REST API). The collected data is stored as raw text data.

[0266] Input: URL or API endpoint of the online platform

[0267] Output: Raw help text data

[0268] Specific behavior: The server accesses the specified URL and parses the page's HTML to extract text information or retrieves JSON-formatted data from an API.

[0269] Step 2: Data analysis and classification

[0270] The server then uses natural language processing (NLP) to analyze the collected raw help information and classify it into specific categories (e.g., account management, payment issues, content management, etc.) using techniques such as tokenization, stemming, and part-of-speech tagging.

[0271] Input: Raw help text data

[0272] Output: Text data classified by category

[0273] What it does: The server uses an NLP library (e.g., spaCy, NLTK) to analyze the meaning of each piece of text and classify it into the appropriate category.

[0274] Step 3: Training the AI ​​model

[0275] The server trains an AI model (e.g., BERT, GPT) based on the classified information. The model uses this information to improve its ability to generate appropriate answers to user questions.

[0276] Input: Categorized text data

[0277] Output: Trained AI model

[0278] Specific operation: The server inputs the classified dataset into the AI ​​model and trains it to optimize the model's parameters.

[0279] Step 4: Receiving the prompt

[0280] Users can use their devices to input prompts to send questions to the system, such as "How do I delete my account?" or "What are the benefits of a premium membership?"

[0281] Input: The prompt text that the user enters

[0282] Output: The prompt sent to the server

[0283] Specific behavior: The user enters a question into the input form, and the content is sent to the server.

[0284] Step 5: Parsing the Question

[0285] The server analyzes the prompt using natural language processing (NLP) to understand the intent of the question.

[0286] Input: prompt statement

[0287] Output: Parsed question intent information

[0288] What it does: The server uses an NLP library to parse the prompt and extract keywords and context from the question.

[0289] Step 6: Generate an answer

[0290] The server searches for relevant data from the collected help information based on the analyzed question intent and generates the best answer, utilizing AI models in this process.

[0291] Input: Parsed question intent information and collected help information

[0292] Output: The generated answer

[0293] How it works: The server searches the database based on the intent of the question and inputs relevant information into the AI ​​model to generate the best answer.

[0294] Step 7: Provide your answers

[0295] The generated answer is sent to the user's terminal and displayed to the user.

[0296] Input: Generated Answer

[0297] Output: The answer displayed on the user's terminal

[0298] Specific operation: The server generates an answer, which is sent to the user's device and displayed on an interface for the user to confirm.

[0299] The system allows users to quickly and accurately obtain information about multiple online platforms.

[0300] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0301] The present invention aims to provide a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms, and to improve the user experience by combining it with an emotion engine that recognizes and responds to user emotions.

[0302] System Overview

[0303] The system is mainly composed of the following four roles (server, terminal, user, and emotion engine).

[0304] 1. The server collects information from the online platform, analyzes it, stores it in a database, receives questions from users, generates answers to those questions, and sends them to the device.

[0305] 2. The terminal provides an interface for the user to enter questions and displays the answers received from the server.

[0306] 3. The user accesses the system through a terminal, enters a question, and obtains the required information.

[0307] 4. The emotion engine recognizes emotions from user input and adjusts the tone and content of responses accordingly.

[0308] Program processing

[0309] 1. Data Collection

[0310] The server collects information from the official help pages of multiple online platforms, using scraping technology and APIs.

[0311] 2. Data Analysis and Classification

[0312] The server analyzes the collected help information using natural language processing (NLP) and classifies each piece of information into specific categories (e.g., listing methods, fees, packaging methods, etc.).

[0313] 3. Training the AI ​​model

[0314] The server uses the classified information to train an AI model, which improves its ability to generate appropriate answers to user questions.

[0315] 4. Receiving and analyzing questions

[0316] When a user uses a device to enter a question, the question is sent to a server, which receives the question and analyzes its intent using natural language processing.

[0317] 5. Emotional Recognition

[0318] The server sends the user's question text to the emotion engine, which analyzes and identifies the user's emotional state. For example, it recognizes that the user is confused based on the word "troubled."

[0319] 6. Answer Generation

[0320] The server searches for relevant information from a database based on the analyzed question and emotion recognition results, and generates an appropriate response, adjusting the tone and content if necessary depending on the emotion.

[0321] 7. Providing answers

[0322] The server sends the generated answer to the terminal, which displays it to the user.

[0323] Specific examples

[0324] Example 1: When a user asks, "What are the fees for Platform A and Platform B?"

[0325] Data collection

[0326] The server collects fee information from the official help pages of Platform A and Platform B.

[0327] Data Analysis and Classification

[0328] The server analyzes the collected fee information and categorizes it for each platform.

[0329] Receiving and parsing questions

[0330] When a user types "What are the fees for Platform A and Platform B?" into a terminal, the question is sent to the server, which analyzes the question and determines that the user is seeking comparative information on fees.

[0331] Emotion recognition

[0332] The server uses an emotion engine to recognize that the user is calm and seeking information.

[0333] Generate answers

[0334] The server searches the database for commission information for Platform A and Platform B and generates a response such as, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[0335] Providing answers

[0336] The generated answers are sent to the device and displayed to the user, who can use them to decide which platform to use.

[0337] Example 2: A user is confused and types, "How do I pack on platform C?"

[0338] Data collection

[0339] The server collects information about packaging methods from the official help page of platform C.

[0340] Data Analysis and Classification

[0341] The server analyzes the collected packaging method information and classifies it into specific steps and recommended packaging materials.

[0342] Receiving and parsing questions

[0343] When a user types "Tell me how to pack on platform C" into a terminal, the question is sent to the server, which analyzes the question and determines that it is asking for information about packing methods.

[0344] Emotion recognition

[0345] The server uses an emotion engine to recognize that the user is confused.

[0346] Generate answers

[0347] The server searches its database for information about the packaging method used by Platform C and generates a specific, friendly answer such as, "The packaging method used by Platform C is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape."

[0348] Providing answers

[0349] The generated answer is sent to the device and displayed to the user, who can use it to properly pack the items.

[0350] The above is a specific embodiment of the system of the present invention. This system allows users to easily obtain cross-sectional information on multiple online platforms and provides optimal answers according to different emotional states. This allows individual sellers to sell more smoothly and eliminates literacy differences when using various platforms.

[0351] The processing flow will be explained below.

[0352] Step 1:

[0353] Finalize the help page

[0354] The server lists the official help page URLs of multiple online platforms.

[0355] Specifically, links to help pages are extracted from the official websites of each platform, and the URLs are registered in a database.

[0356] Step 2:

[0357] Information scraping

[0358] The server accesses the listed URLs, performs scraping, and obtains text information from the pages.

[0359] Specifically, it retrieves HTML content from a specified URL and analyzes and extracts text data from it.

[0360] Step 3:

[0361] Storage in the database

[0362] The server stores the acquired information in a database as structured data, organizing the data into categories that differ for each online platform (e.g., listing method, fees, packaging method, etc.).

[0363] Specifically, through analysis, each item is registered in a database as an independent record.

[0364] Step 4:

[0365] Analysis using natural language processing

[0366] The server analyzes the stored text data using natural language processing (NLP) algorithms to extract key keywords and phrases.

[0367] Specifically, it performs processes such as word segmentation, part-of-speech tagging, and entity recognition to structure the information.

[0368] Step 5:

[0369] Information extraction and classification

[0370] Based on the analysis results, the server classifies and tags information such as the listing process, fees, and packaging methods for each online platform.

[0371] Specifically, the information is assigned to an appropriate category based on the extracted keywords and phrases.

[0372] Step 6:

[0373] Training an AI model

[0374] The server uses the classified information to train the chatbot's AI model.

[0375] Specifically, information from the database is used to train a model using a machine learning algorithm (e.g., deep learning).

[0376] Step 7:

[0377] Chatbot display

[0378] The device displays the chatbot's user interface, which can be provided in a web browser, a mobile app, or a desktop application.

[0379] Specifically, when a user accesses the site, it runs a program that pops up a chat window.

[0380] Step 8:

[0381] User question input

[0382] Users enter questions related to the listing process into the chatbot's input field.

[0383] Specifically, the user enters a question using a keyboard or voice input and sends it through the chatbot interface.

[0384] Step 9:

[0385] Receiving questions

[0386] The server receives the question entered by the user.

[0387] Specifically, it receives data sent via the chat window in real time and begins the analysis process.

[0388] Step 10:

[0389] Intent Analysis

[0390] The server analyzes the received question using a natural language processing algorithm to determine the user's intent.

[0391] Specifically, a question-and-answer model is used to identify the type of information being sought from the question.

[0392] Step 11:

[0393] Emotion recognition

[0394] The server sends the user's question text to the emotion engine, which analyzes and identifies the user's emotional state, for example, recognizing whether the user is confused, angry, or happy from the context of the question.

[0395] Specifically, the emotion engine uses natural language processing technology to identify emotions from text data.

[0396] Step 12:

[0397] Search for related information

[0398] The server searches for relevant information from a database based on the analysis results and emotion recognition results.

[0399] Specifically, the identified categories and keywords are used to quickly search for corresponding data in the database.

[0400] Step 13:

[0401] Answer structure

[0402] The server uses the search results to construct a specific answer to the user's question, adjusting the tone and content if an emotional response is needed.

[0403] Specifically, data is embedded in a response template to generate an easy-to-understand and well-organized response.

[0404] Step 14:

[0405] Submit your answer

[0406] The server then sends the constructed response to the user's terminal.

[0407] Specifically, the generated answer message is sent to the user via the chatbot interface.

[0408] Step 15:

[0409] Show Answers

[0410] The terminal displays the received response on the chatbot's interface.

[0411] Specifically, received messages will be displayed in the chat window so that users can view them immediately.

[0412] These are the detailed processing steps of the system. This system allows users to efficiently obtain information and conduct sales activities using multiple online platforms. Furthermore, the introduction of an emotion engine makes it possible to provide appropriate answers according to the user's emotional state, which is expected to improve the user experience.

[0413] Example 2

[0414] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0415] Previous systems were unable to quickly gather necessary information from multiple online platforms and provide relevant and emotionally appropriate answers to users' questions. Furthermore, they lacked the ability to recognize users' emotions and adjust the tone of their responses, which left the user experience unsatisfactory.

[0416] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0417] In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a machine learning model based on the classified information, means for analyzing a user's question using natural language analysis, means for recognizing emotions from the user's question text, means for generating an appropriate answer based on the analyzed question and the recognized emotion, and means for providing the generated answer to a user terminal. This makes it possible to efficiently collect information from multiple online platforms and provide appropriate and emotionally appropriate answers to user questions.

[0418] "Multiple online platforms" refers to multiple websites or applications that offer different services or functions over the Internet.

[0419] "Help information" refers to the explanations and guidelines provided by each online platform to help users understand how to use the platform and how to troubleshoot problems.

[0420] "Means of collection" refers to the use of web scraping technology or APIs to extract the necessary information from online platforms.

[0421] "Means of analyzing and classifying into specific categories" refers to a method of analyzing collected information using natural language processing technology and organizing it into specific categories such as listing method, fees, and packaging method.

[0422] "Methods for training machine learning models" refers to methods for using classified information to train AI models and improve their performance.

[0423] "Means of analyzing user questions using natural language analysis" refers to a method of analyzing the text entered by the user using natural language processing technology to understand its intent and content.

[0424] "Means for recognizing emotions" refers to technology that analyzes emotions from text entered by the user and identifies psychological states such as joy, anger, sadness, and happiness.

[0425] "Answer generation means" refers to a method for automatically generating an appropriate answer based on the content of the question and the perceived sentiment.

[0426] "Means for providing to the user's device" refers to the method of sending the generated answer to the device used by the user and displaying it on the screen.

[0427] The present invention aims to provide a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms, and to improve the user experience by combining it with an emotion engine that recognizes and responds to user emotions.

[0428] The system is mainly composed of four elements: the server, the terminal, the user, and the emotion engine. Each element is explained in detail below.

[0429] The server is responsible for collecting information from the official help pages of multiple online platforms, analyzing it, and storing it in a database. Scraping tools such as "Beautiful Soup" and "Scrapy" are used for collection, and "Python's NLTK" and "spaCy" are used to analyze the information. After analysis, the data is classified into categories (e.g., listing method, fees, packaging method, etc.) and used as training data for the AI ​​model.

[0430] The server then trains an AI model using machine learning libraries such as TensorFlow and PyTorch. This model is capable of generating appropriate answers to user questions. When a user types a question using a device, the question is sent to the server as an HTTP request. The server then analyzes the intent of the question using TextBlob and Hugging Face Transformers. For example, if a user types "Tell me the fees for platform A and platform B," it is interpreted as a request for information about fees.

[0431] The server also uses an emotion engine to recognize the user's emotions. The emotion engine uses emotion analysis tools such as DeepMoji and GPT-3. For example, it can identify the user's confused emotion from a question containing the word "confused." Based on the recognized emotion, it can adjust the tone and content of the response.

[0432] Once the appropriate answer is generated, the server sends it to the user's device and displays it. For example, if the user requests an answer such as "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price," the server retrieves that information from the database, generates the appropriate answer, and sends it to the device.

[0433] As a concrete example, consider the case where a user is confused and types, "Please tell me how to pack on Platform C." The server collects, analyzes, and classifies information about packing methods from Platform C's official help page. Recognizing the user's confused emotion, the server generates a specific, gentle-toned answer such as, "The packing method on Platform C is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape," and sends it to the device for display.

[0434] As explained above, the present invention allows users to easily obtain cross-sectional information on multiple online platforms and provides optimal answers according to different emotional states, thereby enabling individual sellers to sell more smoothly and bridging the literacy gap when using various platforms.

[0435] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0436] Step 1:

[0437] Data collection

[0438] The server collects information from the official help pages of multiple online platforms. Specifically, it uses scraping tools such as "Beautiful Soup" and "Scrapy" to analyze the HTML code of each platform and extract the necessary data. It receives the URL of each platform as input and obtains the extracted text data as output.

[0439] Step 2:

[0440] Data Analysis and Classification

[0441] The server analyzes the collected text data using natural language processing (NLP) techniques. It uses Python's NLTK or spaCy to tokenize the text and perform grammatical and semantic analysis. The input is the text data obtained in step 1, and the output is data categorized into specific categories (e.g., listing method, handling fee, packaging method, etc.).

[0442] Step 3:

[0443] Training an AI model

[0444] The server trains a machine learning model based on the classified information. It uses machine learning libraries such as TensorFlow and PyTorch to train the AI ​​model. The input is the classified data obtained in step 2, and the output is a trained AI model capable of generating appropriate answers to user questions.

[0445] Step 4:

[0446] Receiving and parsing questions

[0447] The device accepts questions entered by the user in a form. The received questions are sent to the server as HTTP requests. The server then analyzes the intent of the questions using TextBlob and Hugging Face Transformers. The input is the user's question text, and the output is the analyzed intent data of the question.

[0448] Step 5:

[0449] Emotion recognition

[0450] The server sends the parsed question text to an emotion engine to analyze and identify the user's emotion. It uses emotion analysis tools such as "DeepMoji" and "GPT-3." The input is the parsed question intent data, and the output is data indicating the user's emotional state.

[0451] Step 6:

[0452] Generate answers

[0453] The server searches for the appropriate information from a database based on the analyzed question intent data and the recognized emotion data. It generates answers using generative AI models such as GPT-3 and BERT. The input is the question intent data and emotion data, and the output is an answer text adjusted with the appropriate tone.

[0454] Step 7:

[0455] Providing answers

[0456] The server sends the generated answer as an HTTP response to the terminal, which displays the answer in its user interface, using the generated answer text as input and the answer displayed to the user as output.

[0457] The above is the specific flow of the system program processing.

[0458] (Application example 2)

[0459] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0460] In today's complex online environment, it is difficult for users to quickly and accurately obtain help information across multiple online platforms. Security services, in particular, need a system that allows users to easily access information to address specific questions and issues related to digital security. It is also important to improve the user experience by providing appropriate answers based on user sentiment.

[0461] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a natural language processing model based on the classified information, means for analyzing a user's question using natural language analysis, means for generating an appropriate answer to the analyzed question, means for providing the generated answer to the user terminal, means for analyzing the emotional state from the user's input, and means for adjusting the tone and content of the answer based on the emotional state. This allows a user to easily obtain help information from multiple online platforms and receive an appropriate answer according to their emotional state.

[0462] An "online platform" is a foundation for providing services and information over the Internet.

[0463] "Help information" refers to support information that users can refer to when they are in trouble on the online platform.

[0464] "Means of collecting information" refers to the mechanism for obtaining necessary information, such as help information, from online platforms.

[0465] "Means of analyzing information and classifying it into specific categories" refers to the function of analyzing collected information using natural language processing or manually and classifying it by content.

[0466] A "natural language processing model" is an algorithm or machine learning model for understanding, analyzing, and generating natural language used by humans.

[0467] "Means for analyzing user questions using natural language analysis" refers to technology for understanding questions entered by users and analyzing their content.

[0468] The "means for generating an appropriate answer" is a function for providing correct information corresponding to the analyzed question.

[0469] "Means for providing to the user's device" refers to a mechanism for displaying the generated answer on the device used by the user.

[0470] "Means for analyzing emotional state" refers to technology for recognizing emotions from user input and determining that state.

[0471] "Means to adjust the tone and content of responses based on emotional state" refers to a feature that changes the wording and content of responses depending on the user's emotional state.

[0472] The present invention aims to provide a system that allows users to quickly and appropriately obtain help information on multiple online platforms, and to provide an excellent user experience, particularly in the field of security services. The following describes in detail an embodiment of the present invention.

[0473] Overall system configuration

[0474] The system is primarily composed of a server, a device, a user, and an emotion engine. This configuration enables the system to generate appropriate answers to user questions and provide them in a tone and content that reflects the user's emotions.

[0475] Server Roles

[0476] The server includes the following means:

[0477] 1. Information collection methods:

[0478] The server collects help information from multiple online platforms, using scraping techniques (such as the requests library and BeautifulSoup) and APIs.

[0479] 2. Information analysis and classification methods:

[0480] The collected help information is analyzed using natural language processing (NLP) techniques and classified into specific categories (e.g., "antivirus," "firewall settings," "how to use VPN," etc.) using the nltk library.

[0481] 3. Natural Language Processing Model Training Methods:

[0482] The classified information is used to train an AI model (e.g., GPT-3.5-turbo), which improves its ability to generate appropriate answers to user questions.

[0483] 4. Question analysis means:

[0484] When a user uses a terminal to input a question, the question is received by the server and its intent is analyzed using natural language analysis.

[0485] 5. Emotion recognition means:

[0486] The user's question text is sent to the emotion engine to analyze and identify the user's emotional state, using nltk's SentimentIntensityAnalyzer.

[0487] 6. Answer generation means:

[0488] Based on the analyzed question and the results of emotion recognition, it searches for relevant information from a database and generates an answer with a tone and content that suits the user's emotional state.

[0489] 7. Means of providing answers:

[0490] The generated answer is sent to the user's terminal and displayed to the user through the terminal.

[0491] Device Role

[0492] The terminal provides an interface for the user to enter questions and displays answers received from the server.

[0493] User Roles

[0494] Users access the system through a terminal, enter questions, and obtain the required information.

[0495] Specific Examples

[0496] Input prompt example:

[0497] "My VPN connection keeps dropping out. What should I do?" (a confused user)

[0498] "How do I configure the firewall?" (Calm user)

[0499] In response to these prompts, the system generates appropriate answers from collected help information from online platforms and uses an emotion recognition engine to provide answers in a tone and content that matches the user's emotional state.

[0500] Hardware and Software Use

[0501] Hardware:

[0502] Server: A server located in a high-performance data center.

[0503] Devices: smartphones, computers, tablets, etc.

[0504] software:

[0505] Web scraping: requests library, BeautifulSoup.

[0506] Natural Language Processing: nltk library, SentimentIntensityAnalyzer.

[0507] AI model: transformers library, GPT-3.5-turbo.

[0508] The above configuration and functions enable users to efficiently obtain help information on multiple online platforms and receive appropriate answers according to their emotional state.

[0509] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0510] Step 1: Gather information

[0511] The server collects help information from multiple online platforms. It uses the requests library and BeautifulSoup to scrape information from specified URLs or retrieve information using APIs. The input is the URL or API endpoint, and the output is the raw data of the collected help information.

[0512] Step 2: Information analysis and classification

[0513] The server analyzes the collected help information and classifies it into specific categories. The nltk library is used to analyze the collected help information using natural language processing and classify it into categories based on specific keywords or phrases (e.g., "antivirus," "firewall settings," "how to use VPN"). The input is the raw data of the help information collected in step 1, and the output is information organized by category.

[0514] Step 3: Training the model

[0515] The server trains a natural language processing model based on the classified information. Specifically, it uses the GPT-3.5-turbo model from the Transformers library to train the AI ​​model using the collected data as training data. The input is information organized by category, and the output is a trained AI model.

[0516] Step 4: Parsing the Question

[0517] When a user types a question using a terminal, the question is sent to the server. The server analyzes the intent of the question using natural language analysis. The nltk library is used to process the input question and analyze keywords and context. The input is the question typed by the user on the terminal, and the output is the analyzed keywords and context information.

[0518] Step 5: Emotion Recognition

[0519] The server sends the user's question text to the emotion engine to analyze the user's emotional state. It uses nltk's SentimentIntensityAnalyzer to calculate the emotion score of the question. The input is the analyzed question text, and the output is the emotion score (positive, negative, or neutral).

[0520] Step 6: Generate an answer

[0521] The server searches for relevant information from a database based on the analyzed question and the results of emotion recognition, and generates an appropriate answer. Using a generative AI model (GPT-3.5-turbo), it generates a response and adjusts the tone and content according to the emotion score. The input is the keywords and emotion score of the analyzed question, and the output is the adjusted response.

[0522] Step 7: Provide your answers

[0523] The server sends the generated answer to the user's terminal, which then displays it to the user. The input is the generated answer sentence, and the output is the answer displayed on the user's terminal.

[0524] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0525] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0526] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0527] [Second embodiment]

[0528] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0529] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0530] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0531] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0532] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0533] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0534] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0535] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0536] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0537] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0538] In the smart glasses 214, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0539] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0540] The present invention provides a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms.

[0541] System Overview

[0542] The system is mainly composed of the following three roles (server, terminal, and user).

[0543] 1. The server collects information from the online platform, analyzes it, stores it in a database, receives questions from users, generates answers to those questions, and sends them to the device.

[0544] 2. The terminal provides an interface for the user to enter questions and displays the answers received from the server.

[0545] 3. The user accesses the system through a terminal, enters a question, and obtains the required information.

[0546] Program processing

[0547] 1. Data Collection

[0548] The server collects information from the official help pages of multiple online platforms, using scraping technology and APIs.

[0549] 2. Data Analysis and Classification

[0550] The server analyzes the collected help information using natural language processing (NLP) and classifies each piece of information into specific categories (e.g., listing methods, fees, packaging methods, etc.).

[0551] 3. Training the AI ​​model

[0552] The server uses the classified information to train an AI model, which improves its ability to generate appropriate answers to user questions.

[0553] 4. Receiving and analyzing questions

[0554] When a user uses a device to enter a question, the question is sent to a server, which receives the question and analyzes its intent using natural language processing.

[0555] 5. Answer Generation

[0556] The server searches for relevant information from a database based on the parsed question and generates an appropriate answer.

[0557] 6. Providing answers

[0558] The server sends the generated answer to the terminal, which displays it to the user.

[0559] Specific examples

[0560] Example 1: When a user asks, "What are the fees for Platform A and Platform B?"

[0561] Data collection

[0562] The server collects fee information from the official help pages of Platform A and Platform B.

[0563] Data Analysis and Classification

[0564] The server analyzes the collected fee information and categorizes it for each platform.

[0565] Receiving and parsing questions

[0566] When a user types "What are the fees for Platform A and Platform B?" into a terminal, the question is sent to the server, which analyzes the question and determines that the user is seeking comparative information on fees.

[0567] Generate answers

[0568] The server searches the database for commission information for Platform A and Platform B and generates a response such as, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[0569] Providing answers

[0570] The generated answers are sent to the device and displayed to the user, who can use them to decide which platform to use.

[0571] Example 2: When a user asks, "How do I pack on platform C?"

[0572] Data collection

[0573] The server collects information about packaging methods from the official help page of platform C.

[0574] Data Analysis and Classification

[0575] The server analyzes the collected packaging method information and classifies it into specific steps and recommended packaging materials.

[0576] Receiving and parsing questions

[0577] When a user types "Tell me how to pack on platform C" into a terminal, the question is sent to the server, which analyzes the question and determines that it is asking for information about packing methods.

[0578] Generate answers

[0579] The server searches a database for information about platform C's packaging method and generates a specific answer such as "Platform C's packaging method is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape."

[0580] Providing answers

[0581] The generated answer is sent to the device and displayed to the user, who can use it to properly pack the items.

[0582] As described above, by using the system of the present invention, users can easily obtain cross-sectional information on multiple online platforms, which allows individual sellers to sell more smoothly and eliminates literacy gaps when using various platforms.

[0583] The processing flow will be explained below.

[0584] Step 1:

[0585] Finalize the help page

[0586] The server lists the official help page URLs of multiple online platforms.

[0587] Specifically, links to help pages are extracted from the official websites of each platform, and the URLs are registered in a database.

[0588] Step 2:

[0589] Information scraping

[0590] The server accesses the listed URLs, performs scraping, and obtains text information from the pages.

[0591] Specifically, it retrieves HTML content from a specified URL and analyzes and extracts text data from it.

[0592] Step 3:

[0593] Storage in the database

[0594] The server stores the acquired information in a database as structured data, organizing the data into categories that differ for each reuse service (e.g., listing method, handling fee, packaging method, etc.).

[0595] Specifically, through analysis, each item is registered in a database as an independent record.

[0596] Step 4:

[0597] Analysis using natural language processing

[0598] The server analyzes the stored text data using natural language processing (NLP) algorithms to extract key keywords and phrases.

[0599] Specifically, it performs processes such as word segmentation, part-of-speech tagging, and entity recognition to structure the information.

[0600] Step 5:

[0601] Information extraction and classification

[0602] Based on the analysis results, the server classifies and tags information such as the listing process, fees, and packaging methods for each reuse service.

[0603] Specifically, the information is assigned to an appropriate category based on the extracted keywords and phrases.

[0604] Step 6:

[0605] Training an AI model

[0606] The server uses the classified information to train the chatbot's AI model.

[0607] Specifically, information from the database is used to train a model using a machine learning algorithm (e.g., deep learning).

[0608] Step 7:

[0609] Chatbot display

[0610] The device displays the chatbot's user interface, which can be provided in a web browser, a mobile app, or a desktop application.

[0611] Specifically, when a user accesses the site, it runs a program that pops up a chat window.

[0612] Step 8:

[0613] User question input

[0614] Users enter questions related to the listing process into the chatbot's input field.

[0615] Specifically, the user enters a question using a keyboard or voice input and sends it through the chatbot interface.

[0616] Step 9:

[0617] Receiving questions

[0618] The server receives the question entered by the user.

[0619] Specifically, it receives data sent via the chat window in real time and begins the analysis process.

[0620] Step 10:

[0621] Intent Analysis

[0622] The server analyzes the received question using a natural language processing algorithm to determine the user's intent.

[0623] Specifically, a question-and-answer model is used to identify the type of information being sought from the question.

[0624] Step 11:

[0625] Search for related information

[0626] The server searches the database for relevant information based on the analysis results.

[0627] Specifically, the identified categories and keywords are used to quickly search for corresponding data in the database.

[0628] Step 12:

[0629] Answer structure

[0630] The server uses the search results to construct a specific answer to the user's question. If comparative information is required, the server combines data on multiple reuse services to create the answer.

[0631] Specifically, data is embedded in a response template to generate an easy-to-understand and well-organized response.

[0632] Step 13:

[0633] Submit your answer

[0634] The server then sends the constructed response to the user's terminal.

[0635] Specifically, the generated answer message is sent to the user via the chatbot interface.

[0636] Step 14:

[0637] Show Answers

[0638] The terminal displays the received response on the chatbot's interface.

[0639] Specifically, received messages will be displayed in the chat window so that users can view them immediately.

[0640] The above are the detailed processing steps of the system. This system allows users to efficiently obtain information and conduct sales activities utilizing multiple reuse services.

[0641] Example 1

[0642] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0643] In the past, collecting and analyzing auxiliary information from multiple electronic platforms and providing quick and accurate answers to user questions required a lot of time and effort. In particular, advanced technology was required to efficiently collect and properly analyze information from each platform. Furthermore, when a user's question related to a specific platform, it was difficult to accurately generate an answer. Therefore, the present invention aims to solve these problems and enable users to easily obtain information from various platforms.

[0644] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0645] In this invention, the server includes a means for collecting auxiliary information from multiple electronic platforms, a means for analyzing the collected information and classifying it into specific categories, and a means for training a natural language processing algorithm based on the classified information, thereby enabling the server to provide quick and accurate answers to questions entered by users.

[0646] An "electronic platform" is a general term for services and systems provided online, and refers to the infrastructure through which users can obtain information and conduct transactions.

[0647] "Supporting information" refers to all information such as guidelines, FAQs, manuals, etc. that electronic platforms provide to users.

[0648] "Means of collection" refers to the technologies and methods used to obtain auxiliary information from electronic platforms, including the use of scraping technologies and APIs.

[0649] "Means for analyzing and classifying into specific categories" refers to the method of analyzing the collected auxiliary information using data processing techniques and organizing it into relevant categories.

[0650] "Natural language processing algorithms" refers to a set of computational techniques that enable computers to understand and process human language, including, for example, machine learning models and dictionary-based methods for analyzing the meaning of sentences.

[0651] "Training" refers to the way algorithms are used to train AI models based on classified information, allowing the system to generate more accurate answers.

[0652] "Means for analyzing user input" refers to technology that uses natural language processing technology to analyze questions or requests entered by users and understand their intent and content.

[0653] "Answer generation means" refers to techniques or methods for generating appropriate answers based on analyzed user input, including database searches and the use of generative AI models.

[0654] "Means of providing" refers to the interface or communication means for displaying or communicating the generated answer to the user.

[0655] "Scraping technology" refers to the technique of analyzing the HTML structure of a website and extracting specific information programmatically.

[0656] "API" stands for Application Programming Interface, and refers to an interface that allows software to communicate with each other.

[0657] A "library" refers to a collection of code and data for shared use of program functions and data, and natural language processing libraries include existing analysis algorithms and datasets.

[0658] "JSON format" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data exchange format for defining data structures.

[0659] This system collects, analyzes, and classifies auxiliary information from multiple electronic platforms, enabling it to respond quickly and accurately to user inquiries. The system is primarily comprised of three roles: a server, a terminal, and a user.

[0660] server

[0661] The server mainly performs the following tasks:

[0662] 1. Data Collection

[0663] The server uses scraping techniques and APIs to collect auxiliary information from electronic platforms. Specifically, it uses Python's BeautifulSoup library to analyze the HTML structure of web pages and extract the necessary information. It can also use appropriate platform APIs to obtain official information.

[0664] 2. Data Analysis and Classification

[0665] The server analyzes the collected auxiliary information using natural language processing (NLP) techniques, such as SpaCy and NLTK, to analyze the meaning of the text data and organize the information into specific categories (e.g., listing method, handling fee, packaging method).

[0666] 3. Training the AI ​​model

[0667] The server trains an AI model based on the classified information. Machine learning libraries such as Scikit-learn and TensorFlow are used to build the AI ​​model. The training dataset consists of the collected and classified auxiliary information. The trained AI model is then capable of generating appropriate answers to user questions.

[0668] 4. Receiving and analyzing questions

[0669] When a user uses a device to input a question, the question is sent to the server, which uses natural language processing (NLP) to analyze the received question and determine its intent. For example, if a user inputs "What are the fees for platform A and platform B," the server understands that the user is seeking comparative information on fees.

[0670] 5. Answer Generation

[0671] The server searches for relevant information from a database based on the parsed question and generates an appropriate answer for the user, which is then sent to the device in JSON format.

[0672] Terminal

[0673] The terminal provides an interface for users to enter questions and displays the answers received from the server. Specifically, when a user enters a question in a text input field and presses the submit button, the question is sent to the server as an HTTP request. After receiving the answer from the server, it is displayed in an easy-to-understand manner to the user using HTML and CSS.

[0674] user

[0675] Users access the system through their terminals, input questions, and obtain the necessary information. For example, if a user asks, "Tell me how to pack on platform C," the question is analyzed by the server, and relevant information is generated and displayed on the terminal.

[0676] Specific examples

[0677] For example, if a user asks "What are the fees for Platform A and Platform B?", the following will show:

[0678] The user enters a question into the text input field on the device and presses the send button.

[0679] The device sends the question to the server as an HTTP request.

[0680] The server receives the question and analyzes it using natural language processing.

[0681] The server searches the database for the commission information of Platform A and Platform B and generates a response saying, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[0682] The server sends the generated response in JSON format to the terminal.

[0683] The device displays the received response to the user.

[0684] As described above, by using the system of the present invention, users can easily obtain cross-sectional information on multiple electronic platforms, which allows individual sellers to conduct sales activities more smoothly and eliminates literacy differences when using various platforms.

[0685] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0686] Step 1: Data collection

[0687] Input: URLs or API endpoints for multiple Electron platforms

[0688] Specific operation: The server uses Python's BeautifulSoup to retrieve the HTML of each platform's official help page and extract the necessary information. It also retrieves information using the platform's API.

[0689] Data processing: Extract appropriate text data from the acquired HTML or API response.

[0690] Output: Text data of extracted auxiliary information

[0691] Step 2: Data analysis and classification

[0692] Input: Text data obtained in the data collection step

[0693] What it does: The server uses a natural language processing (NLP) library (e.g., SpaCy, NLTK) to parse the text data, extracting keywords from the text and organizing each piece of information into specific categories (e.g., listing method, handling fee, packaging method).

[0694] Data processing: Based on the analysis, the information is classified into categories.

[0695] Output: A set of classified auxiliary information

[0696] Step 3: Training the AI ​​model

[0697] Input: A set of classified auxiliary information

[0698] Specific operation: The server trains an AI model using a machine learning library (e.g., Scikit-learn, TensorFlow). It uses the classification data as training data and learns input and output patterns.

[0699] Data processing: Extract features from input data and train the model.

[0700] Output: Trained AI model

[0701] Step 4: Receiving the question

[0702] Input: User question (text format)

[0703] Specific operation: The user enters a question into the device interface and presses the send button. The device then sends the question to the server as an HTTP request.

[0704] Data processing: None

[0705] Output: The question sent to the server

[0706] Step 5: Parsing the Question

[0707] Input: The question sent to the server

[0708] How it works: The server uses natural language processing (NLP) techniques to analyze the question and determine its intent. For example, if a user types, "What are the fees for platform A and platform B?", the server understands that the user is looking for comparative information on fees.

[0709] Data processing: Based on the analysis results, identify the intent of the question.

[0710] Output: Parsed question intent

[0711] Step 6: Generate an answer

[0712] Input: Parsed question intent

[0713] Specific operation: The server searches the database for relevant information and generates an appropriate answer. For example, it searches the database for information on "commission fees" and generates an answer such as "Platform A's commission fee is 10% of the sales price, and Platform B's commission fee is 12% of the sales price."

[0714] Data processing: Extracting relevant information and structuring answers.

[0715] Output: Generated answer (text format)

[0716] Step 7: Provide your answers

[0717] Input: Generated answer (text format)

[0718] Specific operation: The server generates a response and sends it to the device in JSON format. The device receives the response and displays it in an easy-to-understand manner for the user using HTML and CSS.

[0719] Data processing: Convert JSON format data into HTML format so that it can be displayed.

[0720] Output: The answer shown to the user

[0721] (Application example 1)

[0722] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0723] When using help information on multiple online platforms, users often find it difficult to quickly and accurately obtain the information they need. This is especially true for users without specialized knowledge, as each platform has different specifications and usage methods. Furthermore, existing systems lack the analytical capabilities to match questions with appropriate answers, potentially resulting in reduced user satisfaction. To address these challenges, a system is needed that allows users to efficiently obtain information and provide quick and accurate answers to their questions.

[0724] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0725] In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a natural language processing model based on the classified information, means for sending questions to the system via prompt sentences, and means for searching the collected help information on the server and generating optimal answers for the user terminal, thereby enabling the server to generate quick and accurate answers to user questions using help information from multiple online platforms.

[0726] An "online platform" is a system that provides services and content via the Internet.

[0727] "Help Information" refers to documentation and FAQs to help users better understand how to use and troubleshoot the Platform.

[0728] "Means of collection" refers to the function of obtaining and organizing information from the Internet.

[0729] "Means of analysis" refers to the process of converting collected information into an understandable format.

[0730] "Categorizing" means separating information into themes.

[0731] A "natural language processing model" is a machine learning algorithm for analyzing text data and understanding its meaning.

[0732] A "prompt" is text that the user enters into their system to ask a question or request.

[0733] The "means for sending a question" is a function for sending the user's input to the server.

[0734] The "means for generating the optimal answer" is a process for generating the most appropriate answer to the user's question.

[0735] A "user terminal" is a device through which a user can enter questions and receive answers.

[0736] This invention provides a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms. The system consists of three main components: a server, a terminal, and a user.

[0737] Server Configuration

[0738] 1. Data Collection

[0739] The server collects information from the official help pages of multiple online platforms using scraping technology and APIs.

[0740] 2. Data Analysis and Classification

[0741] The server analyzes the collected help information using natural language processing (NLP) and classifies it into specific categories (e.g., account management, payment issues, content management, etc.).

[0742] 3. Training the AI ​​model

[0743] The classified information is used to train an AI model (e.g., BERT, GPT, etc.), which improves its ability to generate appropriate answers to user questions.

[0744] 4. Receiving the prompt

[0745] Prompts are used to allow users to send questions to the system using their devices, such as "How do I delete my account?" or "What are the benefits of a premium membership?"

[0746] 5. Answer Generation

[0747] After receiving the prompt, the server generates the best answer based on the collected help information, and AI models are heavily utilized in this process.

[0748] 6. Providing answers

[0749] The generated answer is sent to the user's terminal and displayed to the user.

[0750] Device configuration

[0751] 1. User Interface

[0752] It provides a chat-style interface for users to enter questions.

[0753] 2. Data Transmission

[0754] The terminal sends the user's question to the server as a prompt sentence.

[0755] 3. Answer display

[0756] Receives the response sent by the server and displays it to the user.

[0757] User Actions

[0758] 1. Enter your question

[0759] The user uses the terminal to enter a question for help information as a prompt sentence.

[0760] 2. Receiving and Confirming Responses

[0761] The user checks the answer sent by the server and obtains the necessary information. For example, if the user enters the question "How do I delete my account?", the server generates a specific answer such as "Go to the settings screen and select the account deletion option."

[0762] This system allows users to quickly and accurately obtain information about multiple online platforms, significantly reducing the effort required for users to efficiently understand and use different platforms.

[0763] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0764] Step 1: Data collection

[0765] The server collects information from the official help pages of multiple online platforms using scraping techniques (e.g., BeautifulSoup, Scrapy) and APIs (e.g., REST API). The collected data is stored as raw text data.

[0766] Input: URL or API endpoint of the online platform

[0767] Output: Raw help text data

[0768] Specific behavior: The server accesses the specified URL and parses the page's HTML to extract text information or retrieves JSON-formatted data from an API.

[0769] Step 2: Data analysis and classification

[0770] The server then uses natural language processing (NLP) to analyze the collected raw help information and classify it into specific categories (e.g., account management, payment issues, content management, etc.) using techniques such as tokenization, stemming, and part-of-speech tagging.

[0771] Input: Raw help text data

[0772] Output: Text data classified by category

[0773] What it does: The server uses an NLP library (e.g., spaCy, NLTK) to analyze the meaning of each piece of text and classify it into the appropriate category.

[0774] Step 3: Training the AI ​​model

[0775] The server trains an AI model (e.g., BERT, GPT) based on the classified information. The model uses this information to improve its ability to generate appropriate answers to user questions.

[0776] Input: Categorized text data

[0777] Output: Trained AI model

[0778] Specific operation: The server inputs the classified dataset into the AI ​​model and trains it to optimize the model's parameters.

[0779] Step 4: Receiving the prompt

[0780] Users can use their devices to input prompts to send questions to the system, such as "How do I delete my account?" or "What are the benefits of a premium membership?"

[0781] Input: The prompt text that the user enters

[0782] Output: The prompt sent to the server

[0783] Specific behavior: The user enters a question into the input form, and the content is sent to the server.

[0784] Step 5: Parsing the Question

[0785] The server analyzes the prompt using natural language processing (NLP) to understand the intent of the question.

[0786] Input: prompt statement

[0787] Output: Parsed question intent information

[0788] What it does: The server uses an NLP library to parse the prompt and extract keywords and context from the question.

[0789] Step 6: Generate an answer

[0790] The server searches for relevant data from the collected help information based on the analyzed question intent and generates the best answer, utilizing AI models in this process.

[0791] Input: Parsed question intent information and collected help information

[0792] Output: The generated answer

[0793] How it works: The server searches the database based on the intent of the question and inputs relevant information into the AI ​​model to generate the best answer.

[0794] Step 7: Provide your answers

[0795] The generated answer is sent to the user's terminal and displayed to the user.

[0796] Input: Generated Answer

[0797] Output: The answer displayed on the user's terminal

[0798] Specific operation: The server generates an answer, which is sent to the user's device and displayed on an interface for the user to confirm.

[0799] The system allows users to quickly and accurately obtain information about multiple online platforms.

[0800] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0801] The present invention aims to provide a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms, and to improve the user experience by combining it with an emotion engine that recognizes and responds to user emotions.

[0802] System Overview

[0803] The system is mainly composed of the following four roles (server, terminal, user, and emotion engine).

[0804] 1. The server collects information from the online platform, analyzes it, stores it in a database, receives questions from users, generates answers to those questions, and sends them to the device.

[0805] 2. The terminal provides an interface for the user to enter questions and displays the answers received from the server.

[0806] 3. The user accesses the system through a terminal, enters a question, and obtains the required information.

[0807] 4. The emotion engine recognizes emotions from user input and adjusts the tone and content of responses accordingly.

[0808] Program processing

[0809] 1. Data Collection

[0810] The server collects information from the official help pages of multiple online platforms, using scraping technology and APIs.

[0811] 2. Data Analysis and Classification

[0812] The server analyzes the collected help information using natural language processing (NLP) and classifies each piece of information into specific categories (e.g., listing methods, fees, packaging methods, etc.).

[0813] 3. Training the AI ​​model

[0814] The server uses the classified information to train an AI model, which improves its ability to generate appropriate answers to user questions.

[0815] 4. Receiving and analyzing questions

[0816] When a user uses a device to enter a question, the question is sent to a server, which receives the question and analyzes its intent using natural language processing.

[0817] 5. Emotional Recognition

[0818] The server sends the user's question text to the emotion engine, which analyzes and identifies the user's emotional state. For example, it recognizes that the user is confused based on the word "troubled."

[0819] 6. Answer Generation

[0820] The server searches for relevant information from a database based on the analyzed question and emotion recognition results, and generates an appropriate response, adjusting the tone and content if necessary depending on the emotion.

[0821] 7. Providing answers

[0822] The server sends the generated answer to the terminal, which displays it to the user.

[0823] Specific examples

[0824] Example 1: When a user asks, "What are the fees for Platform A and Platform B?"

[0825] Data collection

[0826] The server collects fee information from the official help pages of Platform A and Platform B.

[0827] Data Analysis and Classification

[0828] The server analyzes the collected fee information and categorizes it for each platform.

[0829] Receiving and parsing questions

[0830] When a user types "What are the fees for Platform A and Platform B?" into a terminal, the question is sent to the server, which analyzes the question and determines that the user is seeking comparative information on fees.

[0831] Emotion recognition

[0832] The server uses an emotion engine to recognize that the user is calm and seeking information.

[0833] Generate answers

[0834] The server searches the database for commission information for Platform A and Platform B and generates a response such as, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[0835] Providing answers

[0836] The generated answers are sent to the device and displayed to the user, who can use them to decide which platform to use.

[0837] Example 2: A user is confused and types, "How do I pack on platform C?"

[0838] Data collection

[0839] The server collects information about packaging methods from the official help page of platform C.

[0840] Data Analysis and Classification

[0841] The server analyzes the collected packaging method information and classifies it into specific steps and recommended packaging materials.

[0842] Receiving and parsing questions

[0843] When a user types "Tell me how to pack on platform C" into a terminal, the question is sent to the server, which analyzes the question and determines that it is asking for information about packing methods.

[0844] Emotion recognition

[0845] The server uses an emotion engine to recognize that the user is confused.

[0846] Generate answers

[0847] The server searches its database for information about the packaging method used by Platform C and generates a specific, friendly answer such as, "The packaging method used by Platform C is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape."

[0848] Providing answers

[0849] The generated answer is sent to the device and displayed to the user, who can use it to properly pack the items.

[0850] The above is a specific embodiment of the system of the present invention. This system allows users to easily obtain cross-sectional information on multiple online platforms and provides optimal answers according to different emotional states. This allows individual sellers to sell more smoothly and eliminates literacy differences when using various platforms.

[0851] The processing flow will be explained below.

[0852] Step 1:

[0853] Finalize the help page

[0854] The server lists the official help page URLs of multiple online platforms.

[0855] Specifically, links to help pages are extracted from the official websites of each platform, and the URLs are registered in a database.

[0856] Step 2:

[0857] Information scraping

[0858] The server accesses the listed URLs, performs scraping, and obtains text information from the pages.

[0859] Specifically, it retrieves HTML content from a specified URL and analyzes and extracts text data from it.

[0860] Step 3:

[0861] Storage in the database

[0862] The server stores the acquired information in a database as structured data, organizing the data into categories that differ for each online platform (e.g., listing method, fees, packaging method, etc.).

[0863] Specifically, through analysis, each item is registered in a database as an independent record.

[0864] Step 4:

[0865] Analysis using natural language processing

[0866] The server analyzes the stored text data using natural language processing (NLP) algorithms to extract key keywords and phrases.

[0867] Specifically, it performs processes such as word segmentation, part-of-speech tagging, and entity recognition to structure the information.

[0868] Step 5:

[0869] Information extraction and classification

[0870] Based on the analysis results, the server classifies and tags information such as the listing process, fees, and packaging methods for each online platform.

[0871] Specifically, the information is assigned to an appropriate category based on the extracted keywords and phrases.

[0872] Step 6:

[0873] Training an AI model

[0874] The server uses the classified information to train the chatbot's AI model.

[0875] Specifically, information from the database is used to train a model using a machine learning algorithm (e.g., deep learning).

[0876] Step 7:

[0877] Chatbot display

[0878] The device displays the chatbot's user interface, which can be provided in a web browser, a mobile app, or a desktop application.

[0879] Specifically, when a user accesses the site, it runs a program that pops up a chat window.

[0880] Step 8:

[0881] User question input

[0882] Users enter questions related to the listing process into the chatbot's input field.

[0883] Specifically, the user enters a question using a keyboard or voice input and sends it through the chatbot interface.

[0884] Step 9:

[0885] Receiving questions

[0886] The server receives the question entered by the user.

[0887] Specifically, it receives data sent via the chat window in real time and begins the analysis process.

[0888] Step 10:

[0889] Intent Analysis

[0890] The server analyzes the received question using a natural language processing algorithm to determine the user's intent.

[0891] Specifically, a question-and-answer model is used to identify the type of information being sought from the question.

[0892] Step 11:

[0893] Emotion recognition

[0894] The server sends the user's question text to the emotion engine, which analyzes and identifies the user's emotional state, for example, recognizing whether the user is confused, angry, or happy from the context of the question.

[0895] Specifically, the emotion engine uses natural language processing technology to identify emotions from text data.

[0896] Step 12:

[0897] Search for related information

[0898] The server searches for relevant information from a database based on the analysis results and emotion recognition results.

[0899] Specifically, the identified categories and keywords are used to quickly search for corresponding data in the database.

[0900] Step 13:

[0901] Answer structure

[0902] The server uses the search results to construct a specific answer to the user's question, adjusting the tone and content if an emotional response is needed.

[0903] Specifically, data is embedded in a response template to generate an easy-to-understand and well-organized response.

[0904] Step 14:

[0905] Submit your answer

[0906] The server then sends the constructed response to the user's terminal.

[0907] Specifically, the generated answer message is sent to the user via the chatbot interface.

[0908] Step 15:

[0909] Show Answers

[0910] The terminal displays the received response on the chatbot's interface.

[0911] Specifically, received messages will be displayed in the chat window so that users can view them immediately.

[0912] These are the detailed processing steps of the system. This system allows users to efficiently obtain information and conduct sales activities using multiple online platforms. Furthermore, the introduction of an emotion engine makes it possible to provide appropriate answers according to the user's emotional state, which is expected to improve the user experience.

[0913] Example 2

[0914] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0915] Previous systems were unable to quickly gather necessary information from multiple online platforms and provide relevant and emotionally appropriate answers to users' questions. Furthermore, they lacked the ability to recognize users' emotions and adjust the tone of their responses, which left the user experience unsatisfactory.

[0916] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0917] In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a machine learning model based on the classified information, means for analyzing a user's question using natural language analysis, means for recognizing emotions from the user's question text, means for generating an appropriate answer based on the analyzed question and the recognized emotion, and means for providing the generated answer to a user terminal. This makes it possible to efficiently collect information from multiple online platforms and provide appropriate and emotionally appropriate answers to user questions.

[0918] "Multiple online platforms" refers to multiple websites or applications that offer different services or functions over the Internet.

[0919] "Help information" refers to the explanations and guidelines provided by each online platform to help users understand how to use the platform and how to troubleshoot problems.

[0920] "Means of collection" refers to the use of web scraping technology or APIs to extract the necessary information from online platforms.

[0921] "Means of analyzing and classifying into specific categories" refers to a method of analyzing collected information using natural language processing technology and organizing it into specific categories such as listing method, fees, and packaging method.

[0922] "Methods for training machine learning models" refers to methods for using classified information to train AI models and improve their performance.

[0923] "Means of analyzing user questions using natural language analysis" refers to a method of analyzing the text entered by the user using natural language processing technology to understand its intent and content.

[0924] "Means for recognizing emotions" refers to technology that analyzes emotions from text entered by the user and identifies psychological states such as joy, anger, sadness, and happiness.

[0925] "Answer generation means" refers to a method for automatically generating an appropriate answer based on the content of the question and the perceived sentiment.

[0926] "Means for providing to the user's device" refers to the method of sending the generated answer to the device used by the user and displaying it on the screen.

[0927] The present invention aims to provide a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms, and to improve the user experience by combining it with an emotion engine that recognizes and responds to user emotions.

[0928] The system is mainly composed of four elements: the server, the terminal, the user, and the emotion engine. Each element is explained in detail below.

[0929] The server is responsible for collecting information from the official help pages of multiple online platforms, analyzing it, and storing it in a database. Scraping tools such as "Beautiful Soup" and "Scrapy" are used for collection, and "Python's NLTK" and "spaCy" are used to analyze the information. After analysis, the data is classified into categories (e.g., listing method, fees, packaging method, etc.) and used as training data for the AI ​​model.

[0930] The server then trains an AI model using machine learning libraries such as TensorFlow and PyTorch. This model is capable of generating appropriate answers to user questions. When a user types a question using a device, the question is sent to the server as an HTTP request. The server then analyzes the intent of the question using TextBlob and Hugging Face Transformers. For example, if a user types "Tell me the fees for platform A and platform B," it is interpreted as a request for information about fees.

[0931] The server also uses an emotion engine to recognize the user's emotions. The emotion engine uses emotion analysis tools such as DeepMoji and GPT-3. For example, it can identify the user's confused emotion from a question containing the word "confused." Based on the recognized emotion, it can adjust the tone and content of the response.

[0932] Once the appropriate answer is generated, the server sends it to the user's device and displays it. For example, if the user requests an answer such as "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price," the server retrieves that information from the database, generates the appropriate answer, and sends it to the device.

[0933] As a concrete example, consider the case where a user is confused and types, "Please tell me how to pack on Platform C." The server collects, analyzes, and classifies information about packing methods from Platform C's official help page. Recognizing the user's confused emotion, the server generates a specific, gentle-toned answer such as, "The packing method on Platform C is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape," and sends it to the device for display.

[0934] As explained above, the present invention allows users to easily obtain cross-sectional information on multiple online platforms and provides optimal answers according to different emotional states, thereby enabling individual sellers to sell more smoothly and bridging the literacy gap when using various platforms.

[0935] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0936] Step 1:

[0937] Data collection

[0938] The server collects information from the official help pages of multiple online platforms. Specifically, it uses scraping tools such as "Beautiful Soup" and "Scrapy" to analyze the HTML code of each platform and extract the necessary data. It receives the URL of each platform as input and obtains the extracted text data as output.

[0939] Step 2:

[0940] Data Analysis and Classification

[0941] The server analyzes the collected text data using natural language processing (NLP) techniques. It uses Python's NLTK or spaCy to tokenize the text and perform grammatical and semantic analysis. The input is the text data obtained in step 1, and the output is data categorized into specific categories (e.g., listing method, handling fee, packaging method, etc.).

[0942] Step 3:

[0943] Training an AI model

[0944] The server trains a machine learning model based on the classified information. It uses machine learning libraries such as TensorFlow and PyTorch to train the AI ​​model. The input is the classified data obtained in step 2, and the output is a trained AI model capable of generating appropriate answers to user questions.

[0945] Step 4:

[0946] Receiving and parsing questions

[0947] The device accepts questions entered by the user in a form. The received questions are sent to the server as HTTP requests. The server then analyzes the intent of the questions using TextBlob and Hugging Face Transformers. The input is the user's question text, and the output is the analyzed intent data of the question.

[0948] Step 5:

[0949] Emotion recognition

[0950] The server sends the parsed question text to an emotion engine to analyze and identify the user's emotion. It uses emotion analysis tools such as "DeepMoji" and "GPT-3." The input is the parsed question intent data, and the output is data indicating the user's emotional state.

[0951] Step 6:

[0952] Generate answers

[0953] The server searches for the appropriate information from a database based on the analyzed question intent data and the recognized emotion data. It generates answers using generative AI models such as GPT-3 and BERT. The input is the question intent data and emotion data, and the output is an answer text adjusted with the appropriate tone.

[0954] Step 7:

[0955] Providing answers

[0956] The server sends the generated answer as an HTTP response to the terminal, which displays the answer in its user interface, using the generated answer text as input and the answer displayed to the user as output.

[0957] The above is the specific flow of the system program processing.

[0958] (Application example 2)

[0959] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0960] In today's complex online environment, it is difficult for users to quickly and accurately obtain help information across multiple online platforms. Security services, in particular, need a system that allows users to easily access information to address specific questions and issues related to digital security. It is also important to improve the user experience by providing appropriate answers based on user sentiment.

[0961] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a natural language processing model based on the classified information, means for analyzing a user's question using natural language analysis, means for generating an appropriate answer to the analyzed question, means for providing the generated answer to the user terminal, means for analyzing the emotional state from the user's input, and means for adjusting the tone and content of the answer based on the emotional state. This allows a user to easily obtain help information from multiple online platforms and receive an appropriate answer according to their emotional state.

[0962] An "online platform" is a foundation for providing services and information over the Internet.

[0963] "Help information" refers to support information that users can refer to when they are in trouble on the online platform.

[0964] "Means of collecting information" refers to the mechanism for obtaining necessary information, such as help information, from online platforms.

[0965] "Means of analyzing information and classifying it into specific categories" refers to the function of analyzing collected information using natural language processing or manually and classifying it by content.

[0966] A "natural language processing model" is an algorithm or machine learning model for understanding, analyzing, and generating natural language used by humans.

[0967] "Means for analyzing user questions using natural language analysis" refers to technology for understanding questions entered by users and analyzing their content.

[0968] The "means for generating an appropriate answer" is a function for providing correct information corresponding to the analyzed question.

[0969] "Means for providing to the user's device" refers to a mechanism for displaying the generated answer on the device used by the user.

[0970] "Means for analyzing emotional state" refers to technology for recognizing emotions from user input and determining that state.

[0971] "Means to adjust the tone and content of responses based on emotional state" refers to a feature that changes the wording and content of responses depending on the user's emotional state.

[0972] The present invention aims to provide a system that allows users to quickly and appropriately obtain help information on multiple online platforms, and to provide an excellent user experience, particularly in the field of security services. The following describes in detail an embodiment of the present invention.

[0973] Overall system configuration

[0974] The system is primarily composed of a server, a device, a user, and an emotion engine. This configuration enables the system to generate appropriate answers to user questions and provide them in a tone and content that reflects the user's emotions.

[0975] Server Roles

[0976] The server includes the following means:

[0977] 1. Information collection methods:

[0978] The server collects help information from multiple online platforms, using scraping techniques (such as the requests library and BeautifulSoup) and APIs.

[0979] 2. Information analysis and classification methods:

[0980] The collected help information is analyzed using natural language processing (NLP) techniques and classified into specific categories (e.g., "antivirus," "firewall settings," "how to use VPN," etc.) using the nltk library.

[0981] 3. Natural Language Processing Model Training Methods:

[0982] The classified information is used to train an AI model (e.g., GPT-3.5-turbo), which improves its ability to generate appropriate answers to user questions.

[0983] 4. Question analysis means:

[0984] When a user uses a terminal to input a question, the question is received by the server and its intent is analyzed using natural language analysis.

[0985] 5. Emotion recognition means:

[0986] The user's question text is sent to the emotion engine to analyze and identify the user's emotional state, using nltk's SentimentIntensityAnalyzer.

[0987] 6. Answer generation means:

[0988] Based on the analyzed question and the results of emotion recognition, it searches for relevant information from a database and generates an answer with a tone and content that suits the user's emotional state.

[0989] 7. Means of providing answers:

[0990] The generated answer is sent to the user's terminal and displayed to the user through the terminal.

[0991] Device Role

[0992] The terminal provides an interface for the user to enter questions and displays answers received from the server.

[0993] User Roles

[0994] Users access the system through a terminal, enter questions, and obtain the required information.

[0995] Specific Examples

[0996] Input prompt example:

[0997] "My VPN connection keeps dropping out. What should I do?" (a confused user)

[0998] "How do I configure the firewall?" (Calm user)

[0999] In response to these prompts, the system generates appropriate answers from collected help information from online platforms and uses an emotion recognition engine to provide answers in a tone and content that matches the user's emotional state.

[1000] Hardware and Software Use

[1001] Hardware:

[1002] Server: A server located in a high-performance data center.

[1003] Devices: smartphones, computers, tablets, etc.

[1004] software:

[1005] Web scraping: requests library, BeautifulSoup.

[1006] Natural Language Processing: nltk library, SentimentIntensityAnalyzer.

[1007] AI model: transformers library, GPT-3.5-turbo.

[1008] The above configuration and functions enable users to efficiently obtain help information on multiple online platforms and receive appropriate answers according to their emotional state.

[1009] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1010] Step 1: Gather information

[1011] The server collects help information from multiple online platforms. It uses the requests library and BeautifulSoup to scrape information from specified URLs or retrieve information using APIs. The input is the URL or API endpoint, and the output is the raw data of the collected help information.

[1012] Step 2: Information analysis and classification

[1013] The server analyzes the collected help information and classifies it into specific categories. The nltk library is used to analyze the collected help information using natural language processing and classify it into categories based on specific keywords or phrases (e.g., "antivirus," "firewall settings," "how to use VPN"). The input is the raw data of the help information collected in step 1, and the output is information organized by category.

[1014] Step 3: Training the model

[1015] The server trains a natural language processing model based on the classified information. Specifically, it uses the GPT-3.5-turbo model from the Transformers library to train the AI ​​model using the collected data as training data. The input is information organized by category, and the output is a trained AI model.

[1016] Step 4: Parsing the Question

[1017] When a user types a question using a terminal, the question is sent to the server. The server analyzes the intent of the question using natural language analysis. The nltk library is used to process the input question and analyze keywords and context. The input is the question typed by the user on the terminal, and the output is the analyzed keywords and context information.

[1018] Step 5: Emotion Recognition

[1019] The server sends the user's question text to the emotion engine to analyze the user's emotional state. It uses nltk's SentimentIntensityAnalyzer to calculate the emotion score of the question. The input is the analyzed question text, and the output is the emotion score (positive, negative, or neutral).

[1020] Step 6: Generate an answer

[1021] The server searches for relevant information from a database based on the analyzed question and the results of emotion recognition, and generates an appropriate answer. Using a generative AI model (GPT-3.5-turbo), it generates a response and adjusts the tone and content according to the emotion score. The input is the keywords and emotion score of the analyzed question, and the output is the adjusted response.

[1022] Step 7: Provide your answers

[1023] The server sends the generated answer to the user's terminal, which then displays it to the user. The input is the generated answer sentence, and the output is the answer displayed on the user's terminal.

[1024] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1025] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1026] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1027] [Third embodiment]

[1028] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1029] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1030] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1031] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1032] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1033] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1034] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1035] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1036] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1037] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1038] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1039] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1040] The present invention provides a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms.

[1041] System Overview

[1042] The system is mainly composed of the following three roles (server, terminal, and user).

[1043] 1. The server collects information from the online platform, analyzes it, stores it in a database, receives questions from users, generates answers to those questions, and sends them to the device.

[1044] 2. The terminal provides an interface for the user to enter questions and displays the answers received from the server.

[1045] 3. The user accesses the system through a terminal, enters a question, and obtains the required information.

[1046] Program processing

[1047] 1. Data Collection

[1048] The server collects information from the official help pages of multiple online platforms, using scraping technology and APIs.

[1049] 2. Data Analysis and Classification

[1050] The server analyzes the collected help information using natural language processing (NLP) and classifies each piece of information into specific categories (e.g., listing methods, fees, packaging methods, etc.).

[1051] 3. Training the AI ​​model

[1052] The server uses the classified information to train an AI model, which improves its ability to generate appropriate answers to user questions.

[1053] 4. Receiving and analyzing questions

[1054] When a user uses a device to enter a question, the question is sent to a server, which receives the question and analyzes its intent using natural language processing.

[1055] 5. Answer Generation

[1056] The server searches for relevant information from a database based on the parsed question and generates an appropriate answer.

[1057] 6. Providing answers

[1058] The server sends the generated answer to the terminal, which displays it to the user.

[1059] Specific examples

[1060] Example 1: When a user asks, "What are the fees for Platform A and Platform B?"

[1061] Data collection

[1062] The server collects fee information from the official help pages of Platform A and Platform B.

[1063] Data Analysis and Classification

[1064] The server analyzes the collected fee information and categorizes it for each platform.

[1065] Receiving and parsing questions

[1066] When a user types "What are the fees for Platform A and Platform B?" into a terminal, the question is sent to the server, which analyzes the question and determines that the user is seeking comparative information on fees.

[1067] Generate answers

[1068] The server searches the database for commission information for Platform A and Platform B and generates a response such as, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[1069] Providing answers

[1070] The generated answers are sent to the device and displayed to the user, who can use them to decide which platform to use.

[1071] Example 2: When a user asks, "How do I pack on platform C?"

[1072] Data collection

[1073] The server collects information about packaging methods from the official help page of platform C.

[1074] Data Analysis and Classification

[1075] The server analyzes the collected packaging method information and classifies it into specific steps and recommended packaging materials.

[1076] Receiving and parsing questions

[1077] When a user types "Tell me how to pack on platform C" into a terminal, the question is sent to the server, which analyzes the question and determines that it is asking for information about packing methods.

[1078] Generate answers

[1079] The server searches a database for information about platform C's packaging method and generates a specific answer such as "Platform C's packaging method is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape."

[1080] Providing answers

[1081] The generated answer is sent to the device and displayed to the user, who can use it to properly pack the items.

[1082] As described above, by using the system of the present invention, users can easily obtain cross-sectional information on multiple online platforms, which allows individual sellers to sell more smoothly and eliminates literacy gaps when using various platforms.

[1083] The processing flow will be explained below.

[1084] Step 1:

[1085] Finalize the help page

[1086] The server lists the official help page URLs of multiple online platforms.

[1087] Specifically, links to help pages are extracted from the official websites of each platform, and the URLs are registered in a database.

[1088] Step 2:

[1089] Information scraping

[1090] The server accesses the listed URLs, performs scraping, and obtains text information from the pages.

[1091] Specifically, it retrieves HTML content from a specified URL and analyzes and extracts text data from it.

[1092] Step 3:

[1093] Storage in the database

[1094] The server stores the acquired information in a database as structured data, organizing the data into categories that differ for each reuse service (e.g., listing method, handling fee, packaging method, etc.).

[1095] Specifically, through analysis, each item is registered in a database as an independent record.

[1096] Step 4:

[1097] Analysis using natural language processing

[1098] The server analyzes the stored text data using natural language processing (NLP) algorithms to extract key keywords and phrases.

[1099] Specifically, it performs processes such as word segmentation, part-of-speech tagging, and entity recognition to structure the information.

[1100] Step 5:

[1101] Information extraction and classification

[1102] Based on the analysis results, the server classifies and tags information such as the listing process, fees, and packaging methods for each reuse service.

[1103] Specifically, the information is assigned to an appropriate category based on the extracted keywords and phrases.

[1104] Step 6:

[1105] Training an AI model

[1106] The server uses the classified information to train the chatbot's AI model.

[1107] Specifically, information from the database is used to train a model using a machine learning algorithm (e.g., deep learning).

[1108] Step 7:

[1109] Chatbot display

[1110] The device displays the chatbot's user interface, which can be provided in a web browser, a mobile app, or a desktop application.

[1111] Specifically, when a user accesses the site, it runs a program that pops up a chat window.

[1112] Step 8:

[1113] User question input

[1114] Users enter questions related to the listing process into the chatbot's input field.

[1115] Specifically, the user enters a question using a keyboard or voice input and sends it through the chatbot interface.

[1116] Step 9:

[1117] Receiving questions

[1118] The server receives the question entered by the user.

[1119] Specifically, it receives data sent via the chat window in real time and begins the analysis process.

[1120] Step 10:

[1121] Intent Analysis

[1122] The server analyzes the received question using a natural language processing algorithm to determine the user's intent.

[1123] Specifically, a question-and-answer model is used to identify the type of information being sought from the question.

[1124] Step 11:

[1125] Search for related information

[1126] The server searches the database for relevant information based on the analysis results.

[1127] Specifically, the identified categories and keywords are used to quickly search for corresponding data in the database.

[1128] Step 12:

[1129] Answer structure

[1130] The server uses the search results to construct a specific answer to the user's question. If comparative information is required, the server combines data on multiple reuse services to create the answer.

[1131] Specifically, data is embedded in a response template to generate an easy-to-understand and well-organized response.

[1132] Step 13:

[1133] Submit your answer

[1134] The server then sends the constructed response to the user's terminal.

[1135] Specifically, the generated answer message is sent to the user via the chatbot interface.

[1136] Step 14:

[1137] Show Answers

[1138] The terminal displays the received response on the chatbot's interface.

[1139] Specifically, received messages will be displayed in the chat window so that users can view them immediately.

[1140] The above are the detailed processing steps of the system. This system allows users to efficiently obtain information and conduct sales activities utilizing multiple reuse services.

[1141] Example 1

[1142] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1143] In the past, collecting and analyzing auxiliary information from multiple electronic platforms and providing quick and accurate answers to user questions required a lot of time and effort. In particular, advanced technology was required to efficiently collect and properly analyze information from each platform. Furthermore, when a user's question related to a specific platform, it was difficult to accurately generate an answer. Therefore, the present invention aims to solve these problems and enable users to easily obtain information from various platforms.

[1144] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1145] In this invention, the server includes a means for collecting auxiliary information from multiple electronic platforms, a means for analyzing the collected information and classifying it into specific categories, and a means for training a natural language processing algorithm based on the classified information, thereby enabling the server to provide quick and accurate answers to questions entered by users.

[1146] An "electronic platform" is a general term for services and systems provided online, and refers to the infrastructure through which users can obtain information and conduct transactions.

[1147] "Supporting information" refers to all information such as guidelines, FAQs, manuals, etc. that electronic platforms provide to users.

[1148] "Means of collection" refers to the technologies and methods used to obtain auxiliary information from electronic platforms, including the use of scraping technologies and APIs.

[1149] "Means for analyzing and classifying into specific categories" refers to the method of analyzing the collected auxiliary information using data processing techniques and organizing it into relevant categories.

[1150] "Natural language processing algorithms" refers to a set of computational techniques that enable computers to understand and process human language, including, for example, machine learning models and dictionary-based methods for analyzing the meaning of sentences.

[1151] "Training" refers to the way algorithms are used to train AI models based on classified information, allowing the system to generate more accurate answers.

[1152] "Means for analyzing user input" refers to technology that uses natural language processing technology to analyze questions or requests entered by users and understand their intent and content.

[1153] "Answer generation means" refers to techniques or methods for generating appropriate answers based on analyzed user input, including database searches and the use of generative AI models.

[1154] "Means of providing" refers to the interface or communication means for displaying or communicating the generated answer to the user.

[1155] "Scraping technology" refers to the technique of analyzing the HTML structure of a website and extracting specific information programmatically.

[1156] "API" stands for Application Programming Interface, and refers to an interface that allows software to communicate with each other.

[1157] A "library" refers to a collection of code and data for shared use of program functions and data, and natural language processing libraries include existing analysis algorithms and datasets.

[1158] "JSON format" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data exchange format for defining data structures.

[1159] This system collects, analyzes, and classifies auxiliary information from multiple electronic platforms, enabling it to respond quickly and accurately to user inquiries. The system is primarily comprised of three roles: a server, a terminal, and a user.

[1160] server

[1161] The server mainly performs the following tasks:

[1162] 1. Data Collection

[1163] The server uses scraping techniques and APIs to collect auxiliary information from electronic platforms. Specifically, it uses Python's BeautifulSoup library to analyze the HTML structure of web pages and extract the necessary information. It can also use appropriate platform APIs to obtain official information.

[1164] 2. Data Analysis and Classification

[1165] The server analyzes the collected auxiliary information using natural language processing (NLP) techniques, such as SpaCy and NLTK, to analyze the meaning of the text data and organize the information into specific categories (e.g., listing method, handling fee, packaging method).

[1166] 3. Training the AI ​​model

[1167] The server trains an AI model based on the classified information. Machine learning libraries such as Scikit-learn and TensorFlow are used to build the AI ​​model. The training dataset consists of the collected and classified auxiliary information. The trained AI model is then capable of generating appropriate answers to user questions.

[1168] 4. Receiving and analyzing questions

[1169] When a user uses a device to input a question, the question is sent to the server, which uses natural language processing (NLP) to analyze the received question and determine its intent. For example, if a user inputs "What are the fees for platform A and platform B," the server understands that the user is seeking comparative information on fees.

[1170] 5. Answer Generation

[1171] The server searches for relevant information from a database based on the parsed question and generates an appropriate answer for the user, which is then sent to the device in JSON format.

[1172] Terminal

[1173] The terminal provides an interface for users to enter questions and displays the answers received from the server. Specifically, when a user enters a question in a text input field and presses the submit button, the question is sent to the server as an HTTP request. After receiving the answer from the server, it is displayed in an easy-to-understand manner to the user using HTML and CSS.

[1174] user

[1175] Users access the system through their terminals, input questions, and obtain the necessary information. For example, if a user asks, "Tell me how to pack on platform C," the question is analyzed by the server, and relevant information is generated and displayed on the terminal.

[1176] Specific examples

[1177] For example, if a user asks "What are the fees for Platform A and Platform B?", the following will show:

[1178] The user enters a question into the text input field on the device and presses the send button.

[1179] The device sends the question to the server as an HTTP request.

[1180] The server receives the question and analyzes it using natural language processing.

[1181] The server searches the database for the commission information of Platform A and Platform B and generates a response saying, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[1182] The server sends the generated response in JSON format to the terminal.

[1183] The device displays the received response to the user.

[1184] As described above, by using the system of the present invention, users can easily obtain cross-sectional information on multiple electronic platforms, which allows individual sellers to conduct sales activities more smoothly and eliminates literacy differences when using various platforms.

[1185] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1186] Step 1: Data collection

[1187] Input: URLs or API endpoints for multiple Electron platforms

[1188] Specific operation: The server uses Python's BeautifulSoup to retrieve the HTML of each platform's official help page and extract the necessary information. It also retrieves information using the platform's API.

[1189] Data processing: Extract appropriate text data from the acquired HTML or API response.

[1190] Output: Text data of extracted auxiliary information

[1191] Step 2: Data analysis and classification

[1192] Input: Text data obtained in the data collection step

[1193] What it does: The server uses a natural language processing (NLP) library (e.g., SpaCy, NLTK) to parse the text data, extracting keywords from the text and organizing each piece of information into specific categories (e.g., listing method, handling fee, packaging method).

[1194] Data processing: Based on the analysis, the information is classified into categories.

[1195] Output: A set of classified auxiliary information

[1196] Step 3: Training the AI ​​model

[1197] Input: A set of classified auxiliary information

[1198] Specific operation: The server trains an AI model using a machine learning library (e.g., Scikit-learn, TensorFlow). It uses the classification data as training data and learns input and output patterns.

[1199] Data processing: Extract features from input data and train the model.

[1200] Output: Trained AI model

[1201] Step 4: Receiving the question

[1202] Input: User question (text format)

[1203] Specific operation: The user enters a question into the device interface and presses the send button. The device then sends the question to the server as an HTTP request.

[1204] Data processing: None

[1205] Output: The question sent to the server

[1206] Step 5: Parsing the Question

[1207] Input: The question sent to the server

[1208] How it works: The server uses natural language processing (NLP) techniques to analyze the question and determine its intent. For example, if a user types, "What are the fees for platform A and platform B?", the server understands that the user is looking for comparative information on fees.

[1209] Data processing: Based on the analysis results, identify the intent of the question.

[1210] Output: Parsed question intent

[1211] Step 6: Generate an answer

[1212] Input: Parsed question intent

[1213] Specific operation: The server searches the database for relevant information and generates an appropriate answer. For example, it searches the database for information on "commission fees" and generates an answer such as "Platform A's commission fee is 10% of the sales price, and Platform B's commission fee is 12% of the sales price."

[1214] Data processing: Extracting relevant information and structuring answers.

[1215] Output: Generated answer (text format)

[1216] Step 7: Provide your answers

[1217] Input: Generated answer (text format)

[1218] Specific operation: The server generates a response and sends it to the device in JSON format. The device receives the response and displays it in an easy-to-understand manner for the user using HTML and CSS.

[1219] Data processing: Convert JSON format data into HTML format so that it can be displayed.

[1220] Output: The answer shown to the user

[1221] (Application example 1)

[1222] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1223] When using help information on multiple online platforms, users often find it difficult to quickly and accurately obtain the information they need. This is especially true for users without specialized knowledge, as each platform has different specifications and usage methods. Furthermore, existing systems lack the analytical capabilities to match questions with appropriate answers, potentially resulting in reduced user satisfaction. To address these challenges, a system is needed that allows users to efficiently obtain information and provide quick and accurate answers to their questions.

[1224] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1225] In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a natural language processing model based on the classified information, means for sending questions to the system via prompt sentences, and means for searching the collected help information on the server and generating optimal answers for the user terminal, thereby enabling the server to generate quick and accurate answers to user questions using help information from multiple online platforms.

[1226] An "online platform" is a system that provides services and content via the Internet.

[1227] "Help Information" refers to documentation and FAQs to help users better understand how to use and troubleshoot the Platform.

[1228] "Means of collection" refers to the function of obtaining and organizing information from the Internet.

[1229] "Means of analysis" refers to the process of converting collected information into an understandable format.

[1230] "Categorizing" means separating information into themes.

[1231] A "natural language processing model" is a machine learning algorithm for analyzing text data and understanding its meaning.

[1232] A "prompt" is text that the user enters into their system to ask a question or request.

[1233] The "means for sending a question" is a function for sending the user's input to the server.

[1234] The "means for generating the optimal answer" is a process for generating the most appropriate answer to the user's question.

[1235] A "user terminal" is a device through which a user can enter questions and receive answers.

[1236] This invention provides a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms. The system consists of three main components: a server, a terminal, and a user.

[1237] Server Configuration

[1238] 1. Data Collection

[1239] The server collects information from the official help pages of multiple online platforms using scraping technology and APIs.

[1240] 2. Data Analysis and Classification

[1241] The server analyzes the collected help information using natural language processing (NLP) and classifies it into specific categories (e.g., account management, payment issues, content management, etc.).

[1242] 3. Training the AI ​​model

[1243] The classified information is used to train an AI model (e.g., BERT, GPT, etc.), which improves its ability to generate appropriate answers to user questions.

[1244] 4. Receiving the prompt

[1245] Prompts are used to allow users to send questions to the system using their devices, such as "How do I delete my account?" or "What are the benefits of a premium membership?"

[1246] 5. Answer Generation

[1247] After receiving the prompt, the server generates the best answer based on the collected help information, and AI models are heavily utilized in this process.

[1248] 6. Providing answers

[1249] The generated answer is sent to the user's terminal and displayed to the user.

[1250] Device configuration

[1251] 1. User Interface

[1252] It provides a chat-style interface for users to enter questions.

[1253] 2. Data Transmission

[1254] The terminal sends the user's question to the server as a prompt sentence.

[1255] 3. Answer display

[1256] Receives the response sent by the server and displays it to the user.

[1257] User Actions

[1258] 1. Enter your question

[1259] The user uses the terminal to enter a question for help information as a prompt sentence.

[1260] 2. Receiving and Confirming Responses

[1261] The user checks the answer sent by the server and obtains the necessary information. For example, if the user enters the question "How do I delete my account?", the server generates a specific answer such as "Go to the settings screen and select the account deletion option."

[1262] This system allows users to quickly and accurately obtain information about multiple online platforms, significantly reducing the effort required for users to efficiently understand and use different platforms.

[1263] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1264] Step 1: Data collection

[1265] The server collects information from the official help pages of multiple online platforms using scraping techniques (e.g., BeautifulSoup, Scrapy) and APIs (e.g., REST API). The collected data is stored as raw text data.

[1266] Input: URL or API endpoint of the online platform

[1267] Output: Raw help text data

[1268] Specific behavior: The server accesses the specified URL and parses the page's HTML to extract text information or retrieves JSON-formatted data from an API.

[1269] Step 2: Data analysis and classification

[1270] The server then uses natural language processing (NLP) to analyze the collected raw help information and classify it into specific categories (e.g., account management, payment issues, content management, etc.) using techniques such as tokenization, stemming, and part-of-speech tagging.

[1271] Input: Raw help text data

[1272] Output: Text data classified by category

[1273] What it does: The server uses an NLP library (e.g., spaCy, NLTK) to analyze the meaning of each piece of text and classify it into the appropriate category.

[1274] Step 3: Training the AI ​​model

[1275] The server trains an AI model (e.g., BERT, GPT) based on the classified information. The model uses this information to improve its ability to generate appropriate answers to user questions.

[1276] Input: Categorized text data

[1277] Output: Trained AI model

[1278] Specific operation: The server inputs the classified dataset into the AI ​​model and trains it to optimize the model's parameters.

[1279] Step 4: Receiving the prompt

[1280] Users can use their devices to input prompts to send questions to the system, such as "How do I delete my account?" or "What are the benefits of a premium membership?"

[1281] Input: The prompt text that the user enters

[1282] Output: The prompt sent to the server

[1283] Specific behavior: The user enters a question into the input form, and the content is sent to the server.

[1284] Step 5: Parsing the Question

[1285] The server analyzes the prompt using natural language processing (NLP) to understand the intent of the question.

[1286] Input: prompt statement

[1287] Output: Parsed question intent information

[1288] What it does: The server uses an NLP library to parse the prompt and extract keywords and context from the question.

[1289] Step 6: Generate an answer

[1290] The server searches for relevant data from the collected help information based on the analyzed question intent and generates the best answer, utilizing AI models in this process.

[1291] Input: Parsed question intent information and collected help information

[1292] Output: The generated answer

[1293] How it works: The server searches the database based on the intent of the question and inputs relevant information into the AI ​​model to generate the best answer.

[1294] Step 7: Provide your answers

[1295] The generated answer is sent to the user's terminal and displayed to the user.

[1296] Input: Generated Answer

[1297] Output: The answer displayed on the user's terminal

[1298] Specific operation: The server generates an answer, which is sent to the user's device and displayed on an interface for the user to confirm.

[1299] The system allows users to quickly and accurately obtain information about multiple online platforms.

[1300] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1301] The present invention aims to provide a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms, and to improve the user experience by combining it with an emotion engine that recognizes and responds to user emotions.

[1302] System Overview

[1303] The system is mainly composed of the following four roles (server, terminal, user, and emotion engine).

[1304] 1. The server collects information from the online platform, analyzes it, stores it in a database, receives questions from users, generates answers to those questions, and sends them to the device.

[1305] 2. The terminal provides an interface for the user to enter questions and displays the answers received from the server.

[1306] 3. The user accesses the system through a terminal, enters a question, and obtains the required information.

[1307] 4. The emotion engine recognizes emotions from user input and adjusts the tone and content of responses accordingly.

[1308] Program processing

[1309] 1. Data Collection

[1310] The server collects information from the official help pages of multiple online platforms, using scraping technology and APIs.

[1311] 2. Data Analysis and Classification

[1312] The server analyzes the collected help information using natural language processing (NLP) and classifies each piece of information into specific categories (e.g., listing methods, fees, packaging methods, etc.).

[1313] 3. Training the AI ​​model

[1314] The server uses the classified information to train an AI model, which improves its ability to generate appropriate answers to user questions.

[1315] 4. Receiving and analyzing questions

[1316] When a user uses a device to enter a question, the question is sent to a server, which receives the question and analyzes its intent using natural language processing.

[1317] 5. Emotional Recognition

[1318] The server sends the user's question text to the emotion engine, which analyzes and identifies the user's emotional state. For example, it recognizes that the user is confused based on the word "troubled."

[1319] 6. Answer Generation

[1320] The server searches for relevant information from a database based on the analyzed question and emotion recognition results, and generates an appropriate response, adjusting the tone and content if necessary depending on the emotion.

[1321] 7. Providing answers

[1322] The server sends the generated answer to the terminal, which displays it to the user.

[1323] Specific examples

[1324] Example 1: When a user asks, "What are the fees for Platform A and Platform B?"

[1325] Data collection

[1326] The server collects fee information from the official help pages of Platform A and Platform B.

[1327] Data Analysis and Classification

[1328] The server analyzes the collected fee information and categorizes it for each platform.

[1329] Receiving and parsing questions

[1330] When a user types "What are the fees for Platform A and Platform B?" into a terminal, the question is sent to the server, which analyzes the question and determines that the user is seeking comparative information on fees.

[1331] Emotion recognition

[1332] The server uses an emotion engine to recognize that the user is calm and seeking information.

[1333] Generate answers

[1334] The server searches the database for commission information for Platform A and Platform B and generates a response such as "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[1335] Providing answers

[1336] The generated answers are sent to the device and displayed to the user, who can use them to decide which platform to use.

[1337] Example 2: A user is confused and types, "How do I pack on platform C?"

[1338] Data collection

[1339] The server collects information about packaging methods from the official help page of platform C.

[1340] Data Analysis and Classification

[1341] The server analyzes the collected packaging method information and classifies it into specific steps and recommended packaging materials.

[1342] Receiving and parsing questions

[1343] When a user types "Tell me how to pack on platform C" into a terminal, the question is sent to the server, which analyzes the question and determines that it is asking for information about packing methods.

[1344] Emotion recognition

[1345] The server uses an emotion engine to recognize that the user is confused.

[1346] Generate answers

[1347] The server searches its database for information about the packaging method used by Platform C and generates a specific, friendly answer such as, "The packaging method used by Platform C is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape."

[1348] Providing answers

[1349] The generated answer is sent to the device and displayed to the user, who can use it to properly pack the items.

[1350] The above is a specific embodiment of the system of the present invention. This system allows users to easily obtain cross-sectional information on multiple online platforms and provides optimal answers according to different emotional states. This allows individual sellers to sell more smoothly and eliminates literacy differences when using various platforms.

[1351] The processing flow will be explained below.

[1352] Step 1:

[1353] Finalize the help page

[1354] The server lists the official help page URLs of multiple online platforms.

[1355] Specifically, links to help pages are extracted from the official websites of each platform, and the URLs are registered in a database.

[1356] Step 2:

[1357] Information scraping

[1358] The server accesses the listed URLs, performs scraping, and obtains text information from the pages.

[1359] Specifically, it retrieves HTML content from a specified URL and analyzes and extracts text data from it.

[1360] Step 3:

[1361] Storage in the database

[1362] The server stores the acquired information in a database as structured data, organizing the data into categories that differ for each online platform (e.g., listing method, fees, packaging method, etc.).

[1363] Specifically, through analysis, each item is registered in a database as an independent record.

[1364] Step 4:

[1365] Analysis using natural language processing

[1366] The server analyzes the stored text data using natural language processing (NLP) algorithms to extract key keywords and phrases.

[1367] Specifically, it performs processes such as word segmentation, part-of-speech tagging, and entity recognition to structure the information.

[1368] Step 5:

[1369] Information extraction and classification

[1370] Based on the analysis results, the server classifies and tags information such as the listing process, fees, and packaging methods for each online platform.

[1371] Specifically, the information is assigned to an appropriate category based on the extracted keywords and phrases.

[1372] Step 6:

[1373] Training an AI model

[1374] The server uses the classified information to train the chatbot's AI model.

[1375] Specifically, information from the database is used to train a model using a machine learning algorithm (e.g., deep learning).

[1376] Step 7:

[1377] Chatbot display

[1378] The device displays the chatbot's user interface, which can be provided in a web browser, a mobile app, or a desktop application.

[1379] Specifically, when a user accesses the site, it runs a program that pops up a chat window.

[1380] Step 8:

[1381] User question input

[1382] Users enter questions related to the listing process into the chatbot's input field.

[1383] Specifically, the user enters a question using a keyboard or voice input and sends it through the chatbot interface.

[1384] Step 9:

[1385] Receiving questions

[1386] The server receives the question entered by the user.

[1387] Specifically, it receives data sent via the chat window in real time and begins the analysis process.

[1388] Step 10:

[1389] Intent Analysis

[1390] The server analyzes the received question using a natural language processing algorithm to determine the user's intent.

[1391] Specifically, a question-and-answer model is used to identify the type of information being sought from the question.

[1392] Step 11:

[1393] Emotion recognition

[1394] The server sends the user's question text to the emotion engine, which analyzes and identifies the user's emotional state, for example, recognizing whether the user is confused, angry, or happy from the context of the question.

[1395] Specifically, the emotion engine uses natural language processing technology to identify emotions from text data.

[1396] Step 12:

[1397] Search for related information

[1398] The server searches for relevant information from a database based on the analysis results and emotion recognition results.

[1399] Specifically, the identified categories and keywords are used to quickly search for corresponding data in the database.

[1400] Step 13:

[1401] Answer structure

[1402] The server uses the search results to construct a specific answer to the user's question, adjusting the tone and content if an emotional response is needed.

[1403] Specifically, data is embedded in a response template to generate an easy-to-understand and well-organized response.

[1404] Step 14:

[1405] Submit your answer

[1406] The server then sends the constructed response to the user's terminal.

[1407] Specifically, the generated answer message is sent to the user via the chatbot interface.

[1408] Step 15:

[1409] Show Answers

[1410] The terminal displays the received response on the chatbot's interface.

[1411] Specifically, received messages will be displayed in the chat window so that users can view them immediately.

[1412] These are the detailed processing steps of the system. This system allows users to efficiently obtain information and conduct sales activities using multiple online platforms. Furthermore, the introduction of an emotion engine makes it possible to provide appropriate answers according to the user's emotional state, which is expected to improve the user experience.

[1413] Example 2

[1414] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1415] Previous systems were unable to quickly gather necessary information from multiple online platforms and provide relevant and emotionally appropriate answers to users' questions. Furthermore, they lacked the ability to recognize users' emotions and adjust the tone of their responses, which left the user experience unsatisfactory.

[1416] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1417] In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a machine learning model based on the classified information, means for analyzing a user's question using natural language analysis, means for recognizing emotions from the user's question text, means for generating an appropriate answer based on the analyzed question and the recognized emotion, and means for providing the generated answer to a user terminal. This makes it possible to efficiently collect information from multiple online platforms and provide appropriate and emotionally appropriate answers to user questions.

[1418] "Multiple online platforms" refers to multiple websites or applications that offer different services or functions over the Internet.

[1419] "Help information" refers to the explanations and guidelines provided by each online platform to help users understand how to use the platform and how to troubleshoot problems.

[1420] "Means of collection" refers to the use of web scraping technology or APIs to extract the necessary information from online platforms.

[1421] "Means of analyzing and classifying into specific categories" refers to a method of analyzing collected information using natural language processing technology and organizing it into specific categories such as listing method, fees, and packaging method.

[1422] "Methods for training machine learning models" refers to methods for using classified information to train AI models and improve their performance.

[1423] "Means of analyzing user questions using natural language analysis" refers to a method of analyzing the text entered by the user using natural language processing technology to understand its intent and content.

[1424] "Means for recognizing emotions" refers to technology that analyzes emotions from text entered by the user and identifies psychological states such as joy, anger, sadness, and happiness.

[1425] "Answer generation means" refers to a method for automatically generating an appropriate answer based on the content of the question and the perceived sentiment.

[1426] "Means for providing to the user's device" refers to the method of sending the generated answer to the device used by the user and displaying it on the screen.

[1427] The present invention aims to provide a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms, and to improve the user experience by combining it with an emotion engine that recognizes and responds to user emotions.

[1428] The system is mainly composed of four elements: the server, the terminal, the user, and the emotion engine. Each element is explained in detail below.

[1429] The server is responsible for collecting information from the official help pages of multiple online platforms, analyzing it, and storing it in a database. Scraping tools such as "Beautiful Soup" and "Scrapy" are used for collection, and "Python's NLTK" and "spaCy" are used to analyze the information. After analysis, the data is classified into categories (e.g., listing method, fees, packaging method, etc.) and used as training data for the AI ​​model.

[1430] The server then trains an AI model using machine learning libraries such as TensorFlow and PyTorch. This model is capable of generating appropriate answers to user questions. When a user types a question using a device, the question is sent to the server as an HTTP request. The server then analyzes the intent of the question using TextBlob and Hugging Face Transformers. For example, if a user types "Tell me the fees for platform A and platform B," it is interpreted as a request for information about fees.

[1431] The server also uses an emotion engine to recognize the user's emotions. The emotion engine uses emotion analysis tools such as DeepMoji and GPT-3. For example, it can identify the user's confused emotion from a question containing the word "confused." Based on the recognized emotion, it can adjust the tone and content of the response.

[1432] Once the appropriate answer is generated, the server sends it to the user's device and displays it. For example, if the user requests an answer such as "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price," the server retrieves that information from the database, generates the appropriate answer, and sends it to the device.

[1433] As a concrete example, consider the case where a user is confused and types, "Please tell me how to pack on Platform C." The server collects, analyzes, and classifies information about packing methods from Platform C's official help page. Recognizing the user's confused emotion, the server generates a specific, gentle-toned answer such as, "The packing method on Platform C is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape," and sends it to the device for display.

[1434] As explained above, the present invention allows users to easily obtain cross-sectional information on multiple online platforms and provides optimal answers according to different emotional states, thereby enabling individual sellers to sell more smoothly and bridging the literacy gap when using various platforms.

[1435] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1436] Step 1:

[1437] Data collection

[1438] The server collects information from the official help pages of multiple online platforms. Specifically, it uses scraping tools such as "Beautiful Soup" and "Scrapy" to analyze the HTML code of each platform and extract the necessary data. It receives the URL of each platform as input and obtains the extracted text data as output.

[1439] Step 2:

[1440] Data Analysis and Classification

[1441] The server analyzes the collected text data using natural language processing (NLP) techniques. It uses Python's NLTK or spaCy to tokenize the text and perform grammatical and semantic analysis. The input is the text data obtained in step 1, and the output is data categorized into specific categories (e.g., listing method, handling fee, packaging method, etc.).

[1442] Step 3:

[1443] Training an AI model

[1444] The server trains a machine learning model based on the classified information. It uses machine learning libraries such as TensorFlow and PyTorch to train the AI ​​model. The input is the classified data obtained in step 2, and the output is a trained AI model capable of generating appropriate answers to user questions.

[1445] Step 4:

[1446] Receiving and parsing questions

[1447] The device accepts questions entered by the user in a form. The received questions are sent to the server as HTTP requests. The server then analyzes the intent of the questions using TextBlob and Hugging Face Transformers. The input is the user's question text, and the output is the analyzed intent data of the question.

[1448] Step 5:

[1449] Emotion recognition

[1450] The server sends the parsed question text to an emotion engine to analyze and identify the user's emotion. It uses emotion analysis tools such as "DeepMoji" and "GPT-3." The input is the parsed question intent data, and the output is data indicating the user's emotional state.

[1451] Step 6:

[1452] Generate answers

[1453] The server searches for the appropriate information from a database based on the analyzed question intent data and the recognized emotion data. It generates answers using generative AI models such as GPT-3 and BERT. The input is the question intent data and emotion data, and the output is an answer text adjusted with the appropriate tone.

[1454] Step 7:

[1455] Providing answers

[1456] The server sends the generated answer as an HTTP response to the terminal, which displays the answer in its user interface, using the generated answer text as input and the answer displayed to the user as output.

[1457] The above is the specific flow of the system program processing.

[1458] (Application example 2)

[1459] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1460] In today's complex online environment, it is difficult for users to quickly and accurately obtain help information across multiple online platforms. Security services, in particular, need a system that allows users to easily access information to address specific questions and issues related to digital security. It is also important to improve the user experience by providing appropriate answers based on user sentiment.

[1461] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a natural language processing model based on the classified information, means for analyzing a user's question using natural language analysis, means for generating an appropriate answer to the analyzed question, means for providing the generated answer to the user terminal, means for analyzing the emotional state from the user's input, and means for adjusting the tone and content of the answer based on the emotional state. This allows a user to easily obtain help information from multiple online platforms and receive an appropriate answer according to their emotional state.

[1462] An "online platform" is a foundation for providing services and information over the Internet.

[1463] "Help information" refers to support information that users can refer to when they are in trouble on the online platform.

[1464] "Means of collecting information" refers to the mechanism for obtaining necessary information, such as help information, from online platforms.

[1465] "Means of analyzing information and classifying it into specific categories" refers to the function of analyzing collected information using natural language processing or manually and classifying it by content.

[1466] A "natural language processing model" is an algorithm or machine learning model for understanding, analyzing, and generating natural language used by humans.

[1467] "Means for analyzing user questions using natural language analysis" refers to technology for understanding questions entered by users and analyzing their content.

[1468] The "means for generating an appropriate answer" is a function for providing correct information corresponding to the analyzed question.

[1469] "Means for providing to the user's device" refers to a mechanism for displaying the generated answer on the device used by the user.

[1470] "Means for analyzing emotional state" refers to technology for recognizing emotions from user input and determining that state.

[1471] "Means to adjust the tone and content of responses based on emotional state" refers to a feature that changes the wording and content of responses depending on the user's emotional state.

[1472] The present invention aims to provide a system that allows users to quickly and appropriately obtain help information on multiple online platforms, and to provide an excellent user experience, particularly in the field of security services. The following describes in detail an embodiment of the present invention.

[1473] Overall system configuration

[1474] The system is primarily composed of a server, a device, a user, and an emotion engine. This configuration enables the system to generate appropriate answers to user questions and provide them in a tone and content that reflects the user's emotions.

[1475] Server Roles

[1476] The server includes the following means:

[1477] 1. Information collection methods:

[1478] The server collects help information from multiple online platforms, using scraping techniques (such as the requests library and BeautifulSoup) and APIs.

[1479] 2. Information analysis and classification methods:

[1480] The collected help information is analyzed using natural language processing (NLP) techniques and classified into specific categories (e.g., "antivirus," "firewall settings," "how to use VPN," etc.) using the nltk library.

[1481] 3. Natural Language Processing Model Training Methods:

[1482] The classified information is used to train an AI model (e.g., GPT-3.5-turbo), which improves its ability to generate appropriate answers to user questions.

[1483] 4. Question analysis means:

[1484] When a user uses a terminal to input a question, the question is received by the server and its intent is analyzed using natural language analysis.

[1485] 5. Emotion recognition means:

[1486] The user's question text is sent to the emotion engine to analyze and identify the user's emotional state, using nltk's SentimentIntensityAnalyzer.

[1487] 6. Answer generation means:

[1488] Based on the analyzed question and the results of emotion recognition, it searches for relevant information from a database and generates an answer with a tone and content that suits the user's emotional state.

[1489] 7. Means of providing answers:

[1490] The generated answer is sent to the user's terminal and displayed to the user through the terminal.

[1491] Device Role

[1492] The terminal provides an interface for the user to enter questions and displays answers received from the server.

[1493] User Roles

[1494] Users access the system through a terminal, enter questions, and obtain the required information.

[1495] Specific Examples

[1496] Input prompt example:

[1497] "My VPN connection keeps dropping out. What should I do?" (a confused user)

[1498] "How do I configure the firewall?" (Calm user)

[1499] In response to these prompts, the system generates appropriate answers from collected help information from online platforms and uses an emotion recognition engine to provide answers in a tone and content that matches the user's emotional state.

[1500] Hardware and Software Use

[1501] Hardware:

[1502] Server: A server located in a high-performance data center.

[1503] Devices: smartphones, computers, tablets, etc.

[1504] software:

[1505] Web scraping: requests library, BeautifulSoup.

[1506] Natural Language Processing: nltk library, SentimentIntensityAnalyzer.

[1507] AI model: transformers library, GPT-3.5-turbo.

[1508] The above configuration and functions enable users to efficiently obtain help information on multiple online platforms and receive appropriate answers according to their emotional state.

[1509] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1510] Step 1: Gather information

[1511] The server collects help information from multiple online platforms. It uses the requests library and BeautifulSoup to scrape information from specified URLs or retrieve information using APIs. The input is the URL or API endpoint, and the output is the raw data of the collected help information.

[1512] Step 2: Information analysis and classification

[1513] The server analyzes the collected help information and classifies it into specific categories. The nltk library is used to analyze the collected help information using natural language processing and classify it into categories based on specific keywords or phrases (e.g., "antivirus," "firewall settings," "how to use VPN"). The input is the raw data of the help information collected in step 1, and the output is information organized by category.

[1514] Step 3: Training the model

[1515] The server trains a natural language processing model based on the classified information. Specifically, it uses the GPT-3.5-turbo model from the Transformers library to train the AI ​​model using the collected data as training data. The input is information organized by category, and the output is a trained AI model.

[1516] Step 4: Parsing the Question

[1517] When a user types a question using a terminal, the question is sent to the server. The server analyzes the intent of the question using natural language analysis. The nltk library is used to process the input question and analyze keywords and context. The input is the question typed by the user on the terminal, and the output is the analyzed keywords and context information.

[1518] Step 5: Emotion Recognition

[1519] The server sends the user's question text to the emotion engine to analyze the user's emotional state. It uses nltk's SentimentIntensityAnalyzer to calculate the emotion score of the question. The input is the analyzed question text, and the output is the emotion score (positive, negative, or neutral).

[1520] Step 6: Generate an answer

[1521] The server searches for relevant information from a database based on the analyzed question and the results of emotion recognition, and generates an appropriate answer. Using a generative AI model (GPT-3.5-turbo), it generates a response and adjusts the tone and content according to the emotion score. The input is the keywords and emotion score of the analyzed question, and the output is the adjusted response.

[1522] Step 7: Provide your answers

[1523] The server sends the generated answer to the user's terminal, which then displays it to the user. The input is the generated answer sentence, and the output is the answer displayed on the user's terminal.

[1524] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1525] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1526] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1527] [Fourth embodiment]

[1528] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1529] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1530] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1531] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1532] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1533] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1534] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1535] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1536] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1537] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1538] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1539] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1540] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1541] The present invention provides a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms.

[1542] System Overview

[1543] The system is mainly composed of the following three roles (server, terminal, and user).

[1544] 1. The server collects information from the online platform, analyzes it, stores it in a database, receives questions from users, generates answers to those questions, and sends them to the device.

[1545] 2. The terminal provides an interface for the user to enter questions and displays the answers received from the server.

[1546] 3. The user accesses the system through a terminal, enters a question, and obtains the required information.

[1547] Program processing

[1548] 1. Data Collection

[1549] The server collects information from the official help pages of multiple online platforms, using scraping technology and APIs.

[1550] 2. Data Analysis and Classification

[1551] The server analyzes the collected help information using natural language processing (NLP) and classifies each piece of information into specific categories (e.g., listing methods, fees, packaging methods, etc.).

[1552] 3. Training the AI ​​model

[1553] The server uses the classified information to train an AI model, which improves its ability to generate appropriate answers to user questions.

[1554] 4. Receiving and analyzing questions

[1555] When a user uses a device to enter a question, the question is sent to a server, which receives the question and analyzes its intent using natural language processing.

[1556] 5. Answer Generation

[1557] The server searches for relevant information from a database based on the parsed question and generates an appropriate answer.

[1558] 6. Providing answers

[1559] The server sends the generated answer to the terminal, which displays it to the user.

[1560] Specific examples

[1561] Example 1: When a user asks, "What are the fees for Platform A and Platform B?"

[1562] Data collection

[1563] The server collects fee information from the official help pages of Platform A and Platform B.

[1564] Data Analysis and Classification

[1565] The server analyzes the collected fee information and categorizes it for each platform.

[1566] Receiving and parsing questions

[1567] When a user types "What are the fees for Platform A and Platform B?" into a terminal, the question is sent to the server, which analyzes the question and determines that the user is seeking comparative information on fees.

[1568] Generate answers

[1569] The server searches the database for commission information for Platform A and Platform B and generates a response such as, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[1570] Providing answers

[1571] The generated answers are sent to the device and displayed to the user, who can use them to decide which platform to use.

[1572] Example 2: When a user asks, "How do I pack on platform C?"

[1573] Data collection

[1574] The server collects information about packaging methods from the official help page of platform C.

[1575] Data Analysis and Classification

[1576] The server analyzes the collected packaging method information and classifies it into specific steps and recommended packaging materials.

[1577] Receiving and parsing questions

[1578] When a user types "Tell me how to pack on platform C" into a terminal, the question is sent to the server, which analyzes the question and determines that it is asking for information about packing methods.

[1579] Generate answers

[1580] The server searches a database for information about platform C's packaging method and generates a specific answer such as "Platform C's packaging method is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape."

[1581] Providing answers

[1582] The generated answer is sent to the device and displayed to the user, who can use it to properly pack the items.

[1583] As described above, by using the system of the present invention, users can easily obtain cross-sectional information on multiple online platforms, which allows individual sellers to conduct sales activities more smoothly and eliminates literacy differences when using various platforms.

[1584] The processing flow will be explained below.

[1585] Step 1:

[1586] Finalize the help page

[1587] The server lists the official help page URLs of multiple online platforms.

[1588] Specifically, links to help pages are extracted from the official websites of each platform, and the URLs are registered in a database.

[1589] Step 2:

[1590] Information scraping

[1591] The server accesses the listed URLs, performs scraping, and obtains text information from the pages.

[1592] Specifically, it retrieves HTML content from a specified URL and analyzes and extracts text data from it.

[1593] Step 3:

[1594] Storage in the database

[1595] The server stores the acquired information in a database as structured data, organizing the data into categories that differ for each reuse service (e.g., listing method, handling fee, packaging method, etc.).

[1596] Specifically, through analysis, each item is registered in a database as an independent record.

[1597] Step 4:

[1598] Analysis using natural language processing

[1599] The server analyzes the stored text data using natural language processing (NLP) algorithms to extract key keywords and phrases.

[1600] Specifically, it performs processes such as word segmentation, part-of-speech tagging, and entity recognition to structure the information.

[1601] Step 5:

[1602] Information extraction and classification

[1603] Based on the analysis results, the server classifies and tags information such as the listing process, fees, and packaging methods for each reuse service.

[1604] Specifically, the information is assigned to an appropriate category based on the extracted keywords and phrases.

[1605] Step 6:

[1606] Training an AI model

[1607] The server uses the classified information to train the chatbot's AI model.

[1608] Specifically, information from the database is used to train a model using a machine learning algorithm (e.g., deep learning).

[1609] Step 7:

[1610] Chatbot display

[1611] The device displays the chatbot's user interface, which can be provided in a web browser, a mobile app, or a desktop application.

[1612] Specifically, when a user accesses the site, it runs a program that pops up a chat window.

[1613] Step 8:

[1614] User question input

[1615] Users enter questions related to the listing process into the chatbot's input field.

[1616] Specifically, the user enters a question using a keyboard or voice input and sends it through the chatbot interface.

[1617] Step 9:

[1618] Receiving questions

[1619] The server receives the question entered by the user.

[1620] Specifically, it receives data sent via the chat window in real time and begins the analysis process.

[1621] Step 10:

[1622] Intent Analysis

[1623] The server analyzes the received question using a natural language processing algorithm to determine the user's intent.

[1624] Specifically, a question-and-answer model is used to identify the type of information being sought from the question.

[1625] Step 11:

[1626] Search for related information

[1627] The server searches the database for relevant information based on the analysis results.

[1628] Specifically, the identified categories and keywords are used to quickly search for corresponding data in the database.

[1629] Step 12:

[1630] Answer structure

[1631] The server uses the search results to construct a specific answer to the user's question. If comparative information is required, the server combines data on multiple reuse services to create the answer.

[1632] Specifically, data is embedded in a response template to generate an easy-to-understand and well-organized response.

[1633] Step 13:

[1634] Submit your answer

[1635] The server then sends the constructed response to the user's terminal.

[1636] Specifically, the generated answer message is sent to the user via the chatbot interface.

[1637] Step 14:

[1638] Show Answers

[1639] The terminal displays the received response on the chatbot's interface.

[1640] Specifically, received messages will be displayed in the chat window so that users can view them immediately.

[1641] The above are the detailed processing steps of the system. This system allows users to efficiently obtain information and conduct sales activities utilizing multiple reuse services.

[1642] Example 1

[1643] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1644] In the past, collecting and analyzing auxiliary information from multiple electronic platforms and providing quick and accurate answers to user questions required a lot of time and effort. In particular, advanced technology was required to efficiently collect and properly analyze information from each platform. Furthermore, when a user's question related to a specific platform, it was difficult to accurately generate an answer. Therefore, the present invention aims to solve these problems and enable users to easily obtain information from various platforms.

[1645] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1646] In this invention, the server includes a means for collecting auxiliary information from multiple electronic platforms, a means for analyzing the collected information and classifying it into specific categories, and a means for training a natural language processing algorithm based on the classified information, thereby enabling the server to provide quick and accurate answers to questions entered by users.

[1647] An "electronic platform" is a general term for services and systems provided online, and refers to the infrastructure through which users can obtain information and conduct transactions.

[1648] "Supporting information" refers to all information such as guidelines, FAQs, manuals, etc. that electronic platforms provide to users.

[1649] "Means of collection" refers to the technologies and methods used to obtain auxiliary information from electronic platforms, including the use of scraping technologies and APIs.

[1650] "Means for analyzing and classifying into specific categories" refers to the method of analyzing the collected auxiliary information using data processing techniques and organizing it into relevant categories.

[1651] "Natural language processing algorithms" refers to a set of computational techniques that enable computers to understand and process human language, including, for example, machine learning models and dictionary-based methods for analyzing the meaning of sentences.

[1652] "Training" refers to the way algorithms are used to train AI models based on classified information, allowing the system to generate more accurate answers.

[1653] "Means for analyzing user input" refers to technology that uses natural language processing technology to analyze questions or requests entered by users and understand their intent and content.

[1654] "Answer generation means" refers to techniques or methods for generating appropriate answers based on analyzed user input, including database searches and the use of generative AI models.

[1655] "Means of providing" refers to the interface or communication means for displaying or communicating the generated answer to the user.

[1656] "Scraping technology" refers to the technique of analyzing the HTML structure of a website and extracting specific information programmatically.

[1657] "API" stands for Application Programming Interface, and refers to an interface that allows software to communicate with each other.

[1658] A "library" refers to a collection of code and data for shared use of program functions and data, and natural language processing libraries include existing analysis algorithms and datasets.

[1659] "JSON format" is an abbreviation for JavaScript Object Notation, and refers to a lightweight data exchange format for defining data structures.

[1660] This system collects, analyzes, and classifies auxiliary information from multiple electronic platforms, enabling it to respond quickly and accurately to user inquiries. The system is primarily comprised of three roles: a server, a terminal, and a user.

[1661] server

[1662] The server mainly performs the following tasks:

[1663] 1. Data Collection

[1664] The server uses scraping techniques and APIs to collect auxiliary information from electronic platforms. Specifically, it uses Python's BeautifulSoup library to analyze the HTML structure of web pages and extract the necessary information. It can also use appropriate platform APIs to obtain official information.

[1665] 2. Data Analysis and Classification

[1666] The server analyzes the collected auxiliary information using natural language processing (NLP) techniques, such as SpaCy and NLTK, to analyze the meaning of the text data and organize the information into specific categories (e.g., listing method, handling fee, packaging method).

[1667] 3. Training the AI ​​model

[1668] The server trains an AI model based on the classified information. Machine learning libraries such as Scikit-learn and TensorFlow are used to build the AI ​​model. The training dataset consists of the collected and classified auxiliary information. The trained AI model is then capable of generating appropriate answers to user questions.

[1669] 4. Receiving and analyzing questions

[1670] When a user uses a device to input a question, the question is sent to the server, which uses natural language processing (NLP) to analyze the received question and determine its intent. For example, if a user inputs "What are the fees for platform A and platform B," the server understands that the user is seeking comparative information on fees.

[1671] 5. Answer Generation

[1672] The server searches for relevant information from a database based on the parsed question and generates an appropriate answer for the user, which is then sent to the device in JSON format.

[1673] Terminal

[1674] The terminal provides an interface for users to enter questions and displays the answers received from the server. Specifically, when a user enters a question in a text input field and presses the submit button, the question is sent to the server as an HTTP request. After receiving the answer from the server, it is displayed in an easy-to-understand manner to the user using HTML and CSS.

[1675] user

[1676] Users access the system through their terminals, input questions, and obtain the necessary information. For example, if a user asks, "Tell me how to pack on platform C," the question is analyzed by the server, and relevant information is generated and displayed on the terminal.

[1677] Specific examples

[1678] For example, if a user asks "What are the fees for Platform A and Platform B?", the following will show:

[1679] The user enters a question into the text input field on the device and presses the send button.

[1680] The device sends the question to the server as an HTTP request.

[1681] The server receives the question and analyzes it using natural language processing.

[1682] The server searches the database for the commission information of Platform A and Platform B and generates a response saying, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[1683] The server sends the generated response in JSON format to the terminal.

[1684] The device displays the received response to the user.

[1685] As described above, by using the system of the present invention, users can easily obtain cross-sectional information on multiple electronic platforms, which allows individual sellers to conduct sales activities more smoothly and eliminates literacy differences when using various platforms.

[1686] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1687] Step 1: Data collection

[1688] Input: URLs or API endpoints for multiple Electron platforms

[1689] Specific operation: The server uses Python's BeautifulSoup to retrieve the HTML of each platform's official help page and extract the necessary information. It also retrieves information using the platform's API.

[1690] Data processing: Extract appropriate text data from the acquired HTML or API response.

[1691] Output: Text data of extracted auxiliary information

[1692] Step 2: Data analysis and classification

[1693] Input: Text data obtained in the data collection step

[1694] What it does: The server uses a natural language processing (NLP) library (e.g., SpaCy, NLTK) to parse the text data, extracting keywords from the text and organizing each piece of information into specific categories (e.g., listing method, handling fee, packaging method).

[1695] Data processing: Based on the analysis, the information is classified into categories.

[1696] Output: A set of classified auxiliary information

[1697] Step 3: Training the AI ​​model

[1698] Input: A set of classified auxiliary information

[1699] Specific operation: The server trains an AI model using a machine learning library (e.g., Scikit-learn, TensorFlow). It uses the classification data as training data and learns input and output patterns.

[1700] Data processing: Extract features from input data and train the model.

[1701] Output: Trained AI model

[1702] Step 4: Receiving the question

[1703] Input: User question (text format)

[1704] Specific operation: The user enters a question into the device interface and presses the send button. The device then sends the question to the server as an HTTP request.

[1705] Data processing: None

[1706] Output: The question sent to the server

[1707] Step 5: Parsing the Question

[1708] Input: The question sent to the server

[1709] How it works: The server uses natural language processing (NLP) techniques to analyze the question and determine its intent. For example, if a user types, "What are the fees for platform A and platform B?", the server understands that the user is looking for comparative information on fees.

[1710] Data processing: Based on the analysis results, identify the intent of the question.

[1711] Output: Parsed question intent

[1712] Step 6: Generate an answer

[1713] Input: Parsed question intent

[1714] Specific operation: The server searches the database for relevant information and generates an appropriate answer. For example, it searches the database for information on "commission fees" and generates an answer such as "Platform A's commission fee is 10% of the sales price, and Platform B's commission fee is 12% of the sales price."

[1715] Data processing: Extracting relevant information and structuring answers.

[1716] Output: Generated answer (text format)

[1717] Step 7: Provide your answers

[1718] Input: Generated answer (text format)

[1719] Specific operation: The server generates a response and sends it to the device in JSON format. The device receives the response and displays it in an easy-to-understand manner for the user using HTML and CSS.

[1720] Data processing: Convert JSON format data into HTML format so that it can be displayed.

[1721] Output: The answer shown to the user

[1722] (Application example 1)

[1723] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1724] When using help information on multiple online platforms, users often find it difficult to quickly and accurately obtain the information they need. This is especially true for users without specialized knowledge, as each platform has different specifications and usage methods. Furthermore, existing systems lack the analytical capabilities to match questions with appropriate answers, potentially resulting in reduced user satisfaction. To address these challenges, a system is needed that allows users to efficiently obtain information and provide quick and accurate answers to their questions.

[1725] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1726] In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a natural language processing model based on the classified information, means for sending questions to the system via prompt sentences, and means for searching the collected help information on the server and generating optimal answers for the user terminal, thereby enabling the server to generate quick and accurate answers to user questions using help information from multiple online platforms.

[1727] An "online platform" is a system that provides services and content via the Internet.

[1728] "Help Information" refers to documentation and FAQs to help users better understand how to use and troubleshoot the Platform.

[1729] "Means of collection" refers to the function of obtaining and organizing information from the Internet.

[1730] "Means of analysis" refers to the process of converting collected information into an understandable format.

[1731] "Categorizing" means separating information into themes.

[1732] A "natural language processing model" is a machine learning algorithm for analyzing text data and understanding its meaning.

[1733] A "prompt" is text that the user enters into their system to ask a question or request.

[1734] The "means for sending a question" is a function for sending the user's input to the server.

[1735] The "means for generating the optimal answer" is a process for generating the most appropriate answer to the user's question.

[1736] A "user terminal" is a device through which a user can enter questions and receive answers.

[1737] This invention provides a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms. The system consists of three main components: a server, a terminal, and a user.

[1738] Server Configuration

[1739] 1. Data Collection

[1740] The server collects information from the official help pages of multiple online platforms using scraping technology and APIs.

[1741] 2. Data Analysis and Classification

[1742] The server analyzes the collected help information using natural language processing (NLP) and classifies it into specific categories (e.g., account management, payment issues, content management, etc.).

[1743] 3. Training the AI ​​model

[1744] The classified information is used to train an AI model (e.g., BERT, GPT, etc.), which improves its ability to generate appropriate answers to user questions.

[1745] 4. Receiving the prompt

[1746] Prompts are used to allow users to send questions to the system using their devices, such as "How do I delete my account?" or "What are the benefits of a premium membership?"

[1747] 5. Answer Generation

[1748] After receiving the prompt, the server generates the best answer based on the collected help information, and AI models are heavily utilized in this process.

[1749] 6. Providing answers

[1750] The generated answer is sent to the user's terminal and displayed to the user.

[1751] Device configuration

[1752] 1. User Interface

[1753] It provides a chat-style interface for users to enter questions.

[1754] 2. Data Transmission

[1755] The terminal sends the user's question to the server as a prompt sentence.

[1756] 3. Answer display

[1757] Receives the response sent by the server and displays it to the user.

[1758] User Actions

[1759] 1. Enter your question

[1760] The user uses the terminal to enter a question for help information as a prompt sentence.

[1761] 2. Receiving and Confirming Responses

[1762] The user checks the answer sent by the server and obtains the necessary information. For example, if the user enters the question "How do I delete my account?", the server generates a specific answer such as "Go to the settings screen and select the account deletion option."

[1763] This system allows users to quickly and accurately obtain information about multiple online platforms, significantly reducing the effort required for users to efficiently understand and use different platforms.

[1764] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1765] Step 1: Data collection

[1766] The server collects information from the official help pages of multiple online platforms using scraping techniques (e.g., BeautifulSoup, Scrapy) and APIs (e.g., REST API). The collected data is stored as raw text data.

[1767] Input: URL or API endpoint of the online platform

[1768] Output: Raw help text data

[1769] Specific behavior: The server accesses the specified URL and parses the page's HTML to extract text information or retrieves JSON-formatted data from an API.

[1770] Step 2: Data analysis and classification

[1771] The server then uses natural language processing (NLP) to analyze the collected raw help information and classify it into specific categories (e.g., account management, payment issues, content management, etc.) using techniques such as tokenization, stemming, and part-of-speech tagging.

[1772] Input: Raw help text data

[1773] Output: Text data classified by category

[1774] What it does: The server uses an NLP library (e.g., spaCy, NLTK) to analyze the meaning of each piece of text and classify it into the appropriate category.

[1775] Step 3: Training the AI ​​model

[1776] The server trains an AI model (e.g., BERT, GPT) based on the classified information. The model uses this information to improve its ability to generate appropriate answers to user questions.

[1777] Input: Categorized text data

[1778] Output: Trained AI model

[1779] Specific operation: The server inputs the classified dataset into the AI ​​model and trains it to optimize the model's parameters.

[1780] Step 4: Receiving the prompt

[1781] Users can use their devices to input prompts to send questions to the system, such as "How do I delete my account?" or "What are the benefits of a premium membership?"

[1782] Input: The prompt text that the user enters

[1783] Output: The prompt sent to the server

[1784] Specific behavior: The user enters a question into the input form, and the content is sent to the server.

[1785] Step 5: Parsing the Question

[1786] The server analyzes the prompt using natural language processing (NLP) to understand the intent of the question.

[1787] Input: prompt statement

[1788] Output: Parsed question intent information

[1789] What it does: The server uses an NLP library to parse the prompt and extract keywords and context from the question.

[1790] Step 6: Generate an answer

[1791] The server searches for relevant data from the collected help information based on the analyzed question intent and generates the best answer, utilizing AI models in this process.

[1792] Input: Parsed question intent information and collected help information

[1793] Output: The generated answer

[1794] How it works: The server searches the database based on the intent of the question and inputs relevant information into the AI ​​model to generate the best answer.

[1795] Step 7: Provide your answers

[1796] The generated answer is sent to the user's terminal and displayed to the user.

[1797] Input: Generated Answer

[1798] Output: The answer displayed on the user's terminal

[1799] Specific operation: The server generates an answer, which is sent to the user's device and displayed on an interface for the user to confirm.

[1800] The system allows users to quickly and accurately obtain information about multiple online platforms.

[1801] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1802] The present invention aims to provide a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms, and to improve the user experience by combining it with an emotion engine that recognizes and responds to user emotions.

[1803] System Overview

[1804] The system is mainly composed of the following four roles (server, terminal, user, and emotion engine).

[1805] 1. The server collects information from the online platform, analyzes it, stores it in a database, receives questions from users, generates answers to those questions, and sends them to the device.

[1806] 2. The terminal provides an interface for the user to enter questions and displays the answers received from the server.

[1807] 3. The user accesses the system through a terminal, enters a question, and obtains the required information.

[1808] 4. The emotion engine recognizes emotions from user input and adjusts the tone and content of responses accordingly.

[1809] Program processing

[1810] 1. Data Collection

[1811] The server collects information from the official help pages of multiple online platforms, using scraping technology and APIs.

[1812] 2. Data Analysis and Classification

[1813] The server analyzes the collected help information using natural language processing (NLP) and classifies each piece of information into specific categories (e.g., listing methods, fees, packaging methods, etc.).

[1814] 3. Training the AI ​​model

[1815] The server uses the classified information to train an AI model, which improves its ability to generate appropriate answers to user questions.

[1816] 4. Receiving and analyzing questions

[1817] When a user uses a device to enter a question, the question is sent to a server, which receives the question and analyzes its intent using natural language processing.

[1818] 5. Emotional Recognition

[1819] The server sends the user's question text to the emotion engine, which analyzes and identifies the user's emotional state. For example, it recognizes that the user is confused based on the word "troubled."

[1820] 6. Answer Generation

[1821] The server searches for relevant information from a database based on the analyzed question and emotion recognition results, and generates an appropriate response, adjusting the tone and content if necessary depending on the emotion.

[1822] 7. Providing answers

[1823] The server sends the generated answer to the terminal, which displays it to the user.

[1824] Specific examples

[1825] Example 1: When a user asks, "What are the fees for Platform A and Platform B?"

[1826] Data collection

[1827] The server collects fee information from the official help pages of Platform A and Platform B.

[1828] Data Analysis and Classification

[1829] The server analyzes the collected fee information and categorizes it for each platform.

[1830] Receiving and parsing questions

[1831] When a user types "What are the fees for Platform A and Platform B?" into a terminal, the question is sent to the server, which analyzes the question and determines that the user is seeking comparative information on fees.

[1832] Emotion recognition

[1833] The server uses an emotion engine to recognize that the user is calm and seeking information.

[1834] Generate answers

[1835] The server searches the database for commission information for Platform A and Platform B and generates a response such as, "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price."

[1836] Providing answers

[1837] The generated answers are sent to the device and displayed to the user, who can use them to decide which platform to use.

[1838] Example 2: A user is confused and types, "How do I pack on platform C?"

[1839] Data collection

[1840] The server collects information about packaging methods from the official help page of platform C.

[1841] Data Analysis and Classification

[1842] The server analyzes the collected packaging method information and classifies it into specific steps and recommended packaging materials.

[1843] Receiving and parsing questions

[1844] When a user types "Tell me how to pack on platform C" into a terminal, the question is sent to the server, which analyzes the question and determines that it is asking for information about packing methods.

[1845] Emotion recognition

[1846] The server uses an emotion engine to recognize that the user is confused.

[1847] Generate answers

[1848] The server searches its database for information about the packaging method used by Platform C and generates a specific, friendly answer such as, "The packaging method used by Platform C is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape."

[1849] Providing answers

[1850] The generated answer is sent to the device and displayed to the user, who can use it to properly pack the items.

[1851] The above is a specific embodiment of the system of the present invention. This system allows users to easily obtain cross-sectional information on multiple online platforms and provides optimal answers according to different emotional states. This allows individual sellers to sell more smoothly and eliminates literacy differences when using various platforms.

[1852] The processing flow will be explained below.

[1853] Step 1:

[1854] Finalize the help page

[1855] The server lists the official help page URLs of multiple online platforms.

[1856] Specifically, links to help pages are extracted from the official websites of each platform, and the URLs are registered in a database.

[1857] Step 2:

[1858] Information scraping

[1859] The server accesses the listed URLs, performs scraping, and obtains text information from the pages.

[1860] Specifically, it retrieves HTML content from a specified URL and analyzes and extracts text data from it.

[1861] Step 3:

[1862] Storage in the database

[1863] The server stores the acquired information in a database as structured data, organizing the data into categories that differ for each online platform (e.g., listing method, fees, packaging method, etc.).

[1864] Specifically, through analysis, each item is registered in a database as an independent record.

[1865] Step 4:

[1866] Analysis using natural language processing

[1867] The server analyzes the stored text data using natural language processing (NLP) algorithms to extract key keywords and phrases.

[1868] Specifically, it performs processes such as word segmentation, part-of-speech tagging, and entity recognition to structure the information.

[1869] Step 5:

[1870] Information extraction and classification

[1871] Based on the analysis results, the server classifies and tags information such as the listing process, fees, and packaging methods for each online platform.

[1872] Specifically, the information is assigned to an appropriate category based on the extracted keywords and phrases.

[1873] Step 6:

[1874] Training an AI model

[1875] The server uses the classified information to train the chatbot's AI model.

[1876] Specifically, information from the database is used to train a model using a machine learning algorithm (e.g., deep learning).

[1877] Step 7:

[1878] Chatbot display

[1879] The device displays the chatbot's user interface, which can be provided in a web browser, a mobile app, or a desktop application.

[1880] Specifically, when a user accesses the site, it runs a program that pops up a chat window.

[1881] Step 8:

[1882] User question input

[1883] Users enter questions related to the listing process into the chatbot's input field.

[1884] Specifically, the user enters a question using a keyboard or voice input and sends it through the chatbot interface.

[1885] Step 9:

[1886] Receiving questions

[1887] The server receives the question entered by the user.

[1888] Specifically, it receives data sent via the chat window in real time and begins the analysis process.

[1889] Step 10:

[1890] Intent Analysis

[1891] The server analyzes the received question using a natural language processing algorithm to determine the user's intent.

[1892] Specifically, a question-and-answer model is used to identify the type of information being sought from the question.

[1893] Step 11:

[1894] Emotion recognition

[1895] The server sends the user's question text to the emotion engine, which analyzes and identifies the user's emotional state, for example, recognizing whether the user is confused, angry, or happy from the context of the question.

[1896] Specifically, the emotion engine uses natural language processing technology to identify emotions from text data.

[1897] Step 12:

[1898] Search for related information

[1899] The server searches for relevant information from a database based on the analysis results and emotion recognition results.

[1900] Specifically, the identified categories and keywords are used to quickly search for corresponding data in the database.

[1901] Step 13:

[1902] Answer structure

[1903] The server uses the search results to construct a specific answer to the user's question, adjusting the tone and content if an emotional response is needed.

[1904] Specifically, data is embedded in a response template to generate an easy-to-understand and well-organized response.

[1905] Step 14:

[1906] Submit your answer

[1907] The server then sends the constructed response to the user's terminal.

[1908] Specifically, the generated answer message is sent to the user via the chatbot interface.

[1909] Step 15:

[1910] Show Answers

[1911] The terminal displays the received response on the chatbot's interface.

[1912] Specifically, received messages will be displayed in the chat window so that users can view them immediately.

[1913] These are the detailed processing steps of the system. This system allows users to efficiently obtain information and conduct sales activities using multiple online platforms. Furthermore, the introduction of an emotion engine makes it possible to provide appropriate answers according to the user's emotional state, which is expected to improve the user experience.

[1914] Example 2

[1915] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1916] Previous systems were unable to quickly gather necessary information from multiple online platforms and provide relevant and emotionally appropriate answers to users' questions. Furthermore, they lacked the ability to recognize users' emotions and adjust the tone of their responses, which left the user experience unsatisfactory.

[1917] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1918] In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a machine learning model based on the classified information, means for analyzing a user's question using natural language analysis, means for recognizing emotions from the user's question text, means for generating an appropriate answer based on the analyzed question and the recognized emotion, and means for providing the generated answer to a user terminal. This makes it possible to efficiently collect information from multiple online platforms and provide appropriate and emotionally appropriate answers to user questions.

[1919] "Multiple online platforms" refers to multiple websites or applications that offer different services or functions over the Internet.

[1920] "Help information" refers to the explanations and guidelines provided by each online platform to help users understand how to use the platform and how to troubleshoot problems.

[1921] "Means of collection" refers to the use of web scraping technology or APIs to extract the necessary information from online platforms.

[1922] "Means of analyzing and classifying into specific categories" refers to a method of analyzing collected information using natural language processing technology and organizing it into specific categories such as listing method, fees, and packaging method.

[1923] "Methods for training machine learning models" refers to methods for using classified information to train AI models and improve their performance.

[1924] "Means of analyzing user questions using natural language analysis" refers to a method of analyzing the text entered by the user using natural language processing technology to understand its intent and content.

[1925] "Means for recognizing emotions" refers to technology that analyzes emotions from text entered by the user and identifies psychological states such as joy, anger, sadness, and happiness.

[1926] "Answer generation means" refers to a method for automatically generating an appropriate answer based on the content of the question and the perceived sentiment.

[1927] "Means for providing to the user's device" refers to the method of sending the generated answer to the device used by the user and displaying it on the screen.

[1928] The present invention aims to provide a system that can quickly respond to user questions by collecting, analyzing, and classifying help information from multiple online platforms, and to improve the user experience by combining it with an emotion engine that recognizes and responds to user emotions.

[1929] The system is mainly composed of four elements: the server, the terminal, the user, and the emotion engine. Each element is explained in detail below.

[1930] The server is responsible for collecting information from the official help pages of multiple online platforms, analyzing it, and storing it in a database. Scraping tools such as "Beautiful Soup" and "Scrapy" are used for collection, and "Python's NLTK" and "spaCy" are used to analyze the information. After analysis, the data is classified into categories (e.g., listing method, fees, packaging method, etc.) and used as training data for the AI ​​model.

[1931] The server then trains an AI model using machine learning libraries such as TensorFlow and PyTorch. This model is capable of generating appropriate answers to user questions. When a user types a question using a device, the question is sent to the server as an HTTP request. The server then analyzes the intent of the question using TextBlob and Hugging Face Transformers. For example, if a user types "Tell me the fees for platform A and platform B," it is interpreted as a request for information about fees.

[1932] The server also uses an emotion engine to recognize the user's emotions. The emotion engine uses emotion analysis tools such as DeepMoji and GPT-3. For example, it can identify the user's confused emotion from a question containing the word "confused." Based on the recognized emotion, it can adjust the tone and content of the response.

[1933] Once the appropriate answer is generated, the server sends it to the user's device and displays it. For example, if the user requests an answer such as "Platform A's commission is 10% of the sales price, and Platform B's commission is 12% of the sales price," the server retrieves that information from the database, generates the appropriate answer, and sends it to the device.

[1934] As a concrete example, consider the case where a user is confused and types, "Please tell me how to pack on Platform C." The server collects, analyzes, and classifies information about packing methods from Platform C's official help page. Recognizing the user's confused emotion, the server generates a specific, gentle-toned answer such as, "The packing method on Platform C is to first wrap the product in bubble wrap, then place it in a cardboard box and seal it with the appropriate tape," and sends it to the device for display.

[1935] As explained above, the present invention allows users to easily obtain cross-sectional information on multiple online platforms and provides optimal answers according to different emotional states, thereby enabling individual sellers to sell more smoothly and bridging the literacy gap when using various platforms.

[1936] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1937] Step 1:

[1938] Data collection

[1939] The server collects information from the official help pages of multiple online platforms. Specifically, it uses scraping tools such as "Beautiful Soup" and "Scrapy" to analyze the HTML code of each platform and extract the necessary data. It receives the URL of each platform as input and obtains the extracted text data as output.

[1940] Step 2:

[1941] Data Analysis and Classification

[1942] The server analyzes the collected text data using natural language processing (NLP) techniques. It uses Python's NLTK or spaCy to tokenize the text and perform grammatical and semantic analysis. The input is the text data obtained in step 1, and the output is data categorized into specific categories (e.g., listing method, handling fee, packaging method, etc.).

[1943] Step 3:

[1944] Training an AI model

[1945] The server trains a machine learning model based on the classified information. It uses machine learning libraries such as TensorFlow and PyTorch to train the AI ​​model. The input is the classified data obtained in step 2, and the output is a trained AI model capable of generating appropriate answers to user questions.

[1946] Step 4:

[1947] Receiving and parsing questions

[1948] The device accepts questions entered by the user in a form. The received questions are sent to the server as HTTP requests. The server then analyzes the intent of the questions using TextBlob and Hugging Face Transformers. The input is the user's question text, and the output is the analyzed intent data of the question.

[1949] Step 5:

[1950] Emotion recognition

[1951] The server sends the parsed question text to an emotion engine to analyze and identify the user's emotion. It uses emotion analysis tools such as "DeepMoji" and "GPT-3." The input is the parsed question intent data, and the output is data indicating the user's emotional state.

[1952] Step 6:

[1953] Generate answers

[1954] The server searches for the appropriate information from a database based on the analyzed question intent data and the recognized emotion data. It generates answers using generative AI models such as GPT-3 and BERT. The input is the question intent data and emotion data, and the output is an answer text adjusted with the appropriate tone.

[1955] Step 7:

[1956] Providing answers

[1957] The server sends the generated answer as an HTTP response to the terminal, which displays the answer in its user interface, using the generated answer text as input and the answer displayed to the user as output.

[1958] The above is the specific flow of the system program processing.

[1959] (Application example 2)

[1960] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1961] In today's complex online environment, it is difficult for users to quickly and accurately obtain help information across multiple online platforms. Security services, in particular, need a system that allows users to easily access information to address specific questions and issues related to digital security. It is also important to improve the user experience by providing appropriate answers based on user sentiment.

[1962] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting help information from multiple online platforms, means for analyzing the collected information and classifying it into specific categories, means for training a natural language processing model based on the classified information, means for analyzing a user's question using natural language analysis, means for generating an appropriate answer to the analyzed question, means for providing the generated answer to the user terminal, means for analyzing the emotional state from the user's input, and means for adjusting the tone and content of the answer based on the emotional state. This allows a user to easily obtain help information from multiple online platforms and receive an appropriate answer according to their emotional state.

[1963] An "online platform" is a foundation for providing services and information over the Internet.

[1964] "Help information" refers to support information that users can refer to when they are in trouble on the online platform.

[1965] "Means of collecting information" refers to the mechanism for obtaining necessary information, such as help information, from online platforms.

[1966] "Means of analyzing information and classifying it into specific categories" refers to the function of analyzing collected information using natural language processing or manually and classifying it by content.

[1967] A "natural language processing model" is an algorithm or machine learning model for understanding, analyzing, and generating natural language used by humans.

[1968] "Means for analyzing user questions using natural language analysis" refers to technology for understanding questions entered by users and analyzing their content.

[1969] The "means for generating an appropriate answer" is a function for providing correct information corresponding to the analyzed question.

[1970] "Means for providing to the user's device" refers to a mechanism for displaying the generated answer on the device used by the user.

[1971] "Means for analyzing emotional state" refers to technology for recognizing emotions from user input and determining that state.

[1972] "Means to adjust the tone and content of responses based on emotional state" refers to a feature that changes the wording and content of responses depending on the user's emotional state.

[1973] The present invention aims to provide a system that allows users to quickly and appropriately obtain help information on multiple online platforms, and to provide an excellent user experience, particularly in the field of security services. The following describes in detail an embodiment of the present invention.

[1974] Overall system configuration

[1975] The system is primarily composed of a server, a device, a user, and an emotion engine. This configuration enables the system to generate appropriate answers to user questions and provide them in a tone and content that reflects the user's emotions.

[1976] Server Roles

[1977] The server includes the following means:

[1978] 1. Information collection methods:

[1979] The server collects help information from multiple online platforms, using scraping techniques (such as the requests library and BeautifulSoup) and APIs.

[1980] 2. Information analysis and classification methods:

[1981] The collected help information is analyzed using natural language processing (NLP) techniques and classified into specific categories (e.g., "antivirus," "firewall settings," "how to use VPN," etc.) using the nltk library.

[1982] 3. Natural Language Processing Model Training Methods:

[1983] The classified information is used to train an AI model (e.g., GPT-3.5-turbo), which improves its ability to generate appropriate answers to user questions.

[1984] 4. Question analysis means:

[1985] When a user uses a terminal to input a question, the question is received by the server and its intent is analyzed using natural language analysis.

[1986] 5. Emotion recognition means:

[1987] The user's question text is sent to the emotion engine to analyze and identify the user's emotional state, using nltk's SentimentIntensityAnalyzer.

[1988] 6. Answer generation means:

[1989] Based on the analyzed question and the results of emotion recognition, it searches for relevant information from a database and generates an answer with a tone and content that suits the user's emotional state.

[1990] 7. Means of providing answers:

[1991] The generated answer is sent to the user's terminal and displayed to the user through the terminal.

[1992] Device Role

[1993] The terminal provides an interface for the user to enter questions and displays answers received from the server.

[1994] User Roles

[1995] Users access the system through a terminal, enter questions, and obtain the required information.

[1996] Specific Examples

[1997] Input prompt example:

[1998] "My VPN connection keeps dropping out. What should I do?" (a confused user)

[1999] "How do I configure the firewall?" (Calm user)

[2000] In response to these prompts, the system generates appropriate answers from collected help information from online platforms and uses an emotion recognition engine to provide answers in a tone and content that matches the user's emotional state.

[2001] Hardware and Software Use

[2002] Hardware:

[2003] Server: A server located in a high-performance data center.

[2004] Devices: smartphones, computers, tablets, etc.

[2005] software:

[2006] Web scraping: requests library, BeautifulSoup.

[2007] Natural Language Processing: nltk library, SentimentIntensityAnalyzer.

[2008] AI model: transformers library, GPT-3.5-turbo.

[2009] The above configuration and functions enable users to efficiently obtain help information on multiple online platforms and receive appropriate answers according to their emotional state.

[2010] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2011] Step 1: Gather information

[2012] The server collects help information from multiple online platforms. It uses the requests library and BeautifulSoup to scrape information from specified URLs or retrieve information using APIs. The input is the URL or API endpoint, and the output is the raw data of the collected help information.

[2013] Step 2: Information analysis and classification

[2014] The server analyzes the collected help information and classifies it into specific categories. The nltk library is used to analyze the collected help information using natural language processing and classify it into categories based on specific keywords or phrases (e.g., "antivirus," "firewall settings," "how to use VPN"). The input is the raw data of the help information collected in step 1, and the output is information organized by category.

[2015] Step 3: Training the model

[2016] The server trains a natural language processing model based on the classified information. Specifically, it uses the GPT-3.5-turbo model from the Transformers library to train the AI ​​model using the collected data as training data. The input is information organized by category, and the output is a trained AI model.

[2017] Step 4: Parsing the Question

[2018] When a user types a question using a terminal, the question is sent to the server. The server analyzes the intent of the question using natural language analysis. The nltk library is used to process the input question and analyze keywords and context. The input is the question typed by the user on the terminal, and the output is the analyzed keywords and context information.

[2019] Step 5: Emotion Recognition

[2020] The server sends the user's question text to the emotion engine to analyze the user's emotional state. It uses nltk's SentimentIntensityAnalyzer to calculate the emotion score of the question. The input is the analyzed question text, and the output is the emotion score (positive, negative, or neutral).

[2021] Step 6: Generate an answer

[2022] The server searches for relevant information from a database based on the analyzed question and the results of emotion recognition, and generates an appropriate answer. Using a generative AI model (GPT-3.5-turbo), it generates a response and adjusts the tone and content according to the emotion score. The input is the keywords and emotion score of the analyzed question, and the output is the adjusted response.

[2023] Step 7: Provide your answers

[2024] The server sends the generated answer to the user's terminal, which then displays it to the user. The input is the generated answer sentence, and the output is the answer displayed on the user's terminal.

[2025] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2026] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[2027] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[2028] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2029] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2030] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2031] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2032] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2033] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2034] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2035] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2036] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2037] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2038] 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.

[2039] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2040] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2041] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2042] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[2043] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[2044] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[2045] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[2046] The following is further disclosed regarding the above embodiment.

[2047] (Claim 1)

[2048] A means of collecting help information from multiple online platforms;

[2049] A means of analyzing and classifying the collected information into specific categories;

[2050] A means for training a natural language processing model based on the classified information;

[2051] a means for analyzing a user's question using natural language analysis;

[2052] a means for generating appropriate answers to the parsed questions;

[2053] means for providing the generated answer to a user terminal;

[2054] A system including:

[2055] (Claim 2)

[2056] 10. The system of claim 1, further comprising means for generating comparative information about a plurality of online platforms in response to a user query.

[2057] (Claim 3)

[2058] 2. The system according to claim 1, further comprising means for extracting specific keywords from the collected help information and managing the corresponding information in a database.

[2059] "Example 1"

[2060] (Claim 1)

[2061] a means for collecting auxiliary information from multiple electronic platforms;

[2062] A means of analyzing and classifying the collected information into specific categories;

[2063] A means for training a natural language processing algorithm based on the classified information;

[2064] means for analyzing user input using natural language analysis;

[2065] means for generating an appropriate answer for the parsed input;

[2066] means for providing the generated answer to a user device;

[2067] The means by which information is collected using scraping technology or APIs;

[2068] A means of using libraries for natural language processing,

[2069] A means to send the generated answer in JSON format, and

[2070] a means for providing an interface on a user's device;

[2071] A system including:

[2072] (Claim 2)

[2073] 10. The system of claim 1, further comprising means for generating comparative information regarding a plurality of electronic platforms in response to a user input.

[2074] (Claim 3)

[2075] 2. The system according to claim 1, further comprising means for extracting specific words and phrases from the collected auxiliary information and managing the corresponding information in data storage.

[2076] "Application Example 1"

[2077] (Claim 1)

[2078] A means of collecting help information from multiple online platforms;

[2079] A means of analyzing and classifying the collected information into specific categories;

[2080] A means for training a natural language processing model based on the classified information;

[2081] a means for analyzing a user's question using natural language analysis;

[2082] a means for generating appropriate answers to the parsed questions;

[2083] means for providing the generated answer to a user terminal;

[2084] means for sending questions to the system via prompts;

[2085] A means for searching the collected help information on the server and generating the most appropriate answer for the user's device;

[2086] A system including:

[2087] (Claim 2)

[2088] 10. The system of claim 1, further comprising means for generating fast and accurate answers to user questions using help information from multiple online platforms.

[2089] (Claim 3)

[2090] The system of claim 1 includes a means for performing advanced analysis of a user's question using a generative AI model based on the collected help information and generating an optimal answer based on a prompt sentence.

[2091] "Example 2: Combining Emotion Engines"

[2092] (Claim 1)

[2093] A means of collecting help information from multiple online platforms;

[2094] A means of analyzing and classifying the collected information into specific categories;

[2095] A means for training a machine learning model based on the classified information;

[2096] a means for analyzing a user's question using natural language analysis;

[2097] a means of recognizing emotions from user question text;

[2098] means for generating an appropriate answer based on the analyzed question and the recognized sentiment;

[2099] means for providing the generated answer to a user terminal;

[2100] A system including:

[2101] (Claim 2)

[2102] 10. The system of claim 1, further comprising means for generating comparative information about a plurality of online platforms in response to a user query.

[2103] (Claim 3)

[2104] 2. The system according to claim 1, further comprising means for extracting specific keywords from the collected help information and managing the corresponding information in a database.

[2105] "Application example 2 when combining emotion engines"

[2106] (Claim 1)

[2107] A means of collecting help information from multiple online platforms;

[2108] A means of analyzing and classifying the collected information into specific categories;

[2109] A means for training a natural language processing model based on the classified information;

[2110] a means for analyzing a user's question using natural language analysis;

[2111] a means for generating appropriate answers to the parsed questions;

[2112] means for providing the generated answer to a user terminal;

[2113] a means for analyzing an emotional state from a user's input;

[2114] a means of adjusting the tone and content of responses based on emotional state;

[2115] A system including:

[2116] (Claim 2)

[2117] 10. The system of claim 1, further comprising means for generating comparative information about a plurality of online platforms in response to a user query.

[2118] (Claim 3)

[2119] 2. The system according to claim 1, further comprising means for extracting specific keywords from the collected help information and managing the corresponding information in a database. [Explanation of symbols]

[2120] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting help information from multiple online platforms; A means of analyzing and classifying the collected information into specific categories; A means for training a natural language processing model based on the classified information; a means for analyzing a user's question using natural language analysis; a means for generating appropriate answers to the parsed questions; means for providing the generated answer to a user terminal; A system including:

2. 10. The system of claim 1, further comprising means for generating comparative information regarding a plurality of online platforms in response to a user query.

3. 2. The system according to claim 1, further comprising means for extracting specific keywords from the collected help information and managing the corresponding information in a database.

Images