system

Abstract

We provide the system. [Solution] A means for building a knowledge database by learning legal data using a generative artificial intelligence model, A means of receiving user queries regarding laws and regulations, A means of referencing appropriate legal provisions from a knowledge database based on a query and generating a logical response, A means of providing the generated answer to the user, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, the method including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] It is very difficult for enterprises, especially small and medium-sized enterprises, to understand and appropriately respond to the complex articles and procedures of the Company Law. As a result, many enterprises may increase the labor, time, and costs related to compliance, and may reduce their business efficiency. In particular, there is a vast amount of legal texts, cases, and opinions of experts, and it is difficult even for the legal departments with specialized knowledge to comprehensively understand them and apply them in practice. Therefore, there is a need to provide a system that can efficiently process information related to the Company Law and enable enterprises to respond legally quickly and appropriately.

Means for Solving the Problems

[0005] The present invention solves the above problems by the following means.

[0006] This proposal proposes a system that includes means for building a knowledge database by learning legal data using a generative artificial intelligence model, means for receiving legal queries from users, means for referencing appropriate legal provisions from the knowledge database based on the queries and generating logical answers, and means for providing the generated answers to users.

[0007] Furthermore, by having the means to collect and format legal texts, case studies, and expert opinions as legal data, it becomes possible to build a more accurate and comprehensive knowledge database. In addition, by adding a means to analyze user queries and provide responses in real time, it enables rapid legal response.

[0008] This allows companies to easily understand complex information related to corporate law and respond to legal matters efficiently.

[0009] A "generative artificial intelligence model" is a type of artificial intelligence that learns from vast amounts of data and has the ability to automatically generate text and information.

[0010] "Legal data" refers to a collection of information including laws, ordinances, regulations, and related texts, case studies, and expert opinions.

[0011] A "knowledge database" refers to a database that aggregates and makes searchable knowledge related to a specific area or field of expertise.

[0012] A "query" is a question or inquiry that a user enters into a system, a request to obtain specific information.

[0013] "Legal provisions" refer to the specific textual parts of articles and regulations written based on laws and regulations.

[0014] A "logical response" refers to a logically consistent, reasonable, and convincing answer based on the entered query.

[0015] "Legal response" refers to the actions and procedures necessary for companies and individuals to comply with laws and regulations.

[0016] "Shaping" refers to the process of converting the acquired data into a more user-friendly format based on certain rules.

[0017] "Real-time response" means that an immediate response is given to a query from a user.

Brief Description of Drawings

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

Mode for Carrying Out the Invention

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

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

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

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

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

[0024] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0025] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0026] [First Embodiment]

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

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

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

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

[0031] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0032] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0033] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

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

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

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

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

[0039] This invention is a system that utilizes generative AI to streamline the complex provisions and procedures of company law, thereby supporting companies' legal compliance. The following describes how the server, terminal, and user entities cooperate to implement the system.

[0040] Server-side embodiment

[0041] The server first collects and formats legal data. This legal data includes legal text, case studies, and expert opinions, which are then formatted appropriately. The formatted data is then aggregated into a knowledge database using a generative instinct artificial intelligence model.

[0042] Next, the server trains a generative artificial intelligence model. This training is performed by training a generative AI model, such as a multi-layer neural network, using the collected legal data. After training, the model will have the ability to extract semantic information from the knowledge database and generate answers to queries.

[0043] The server also receives and analyzes queries from users. These queries relate to company law and are submitted by users in the form of specific questions. Based on the queries, the server searches its knowledge database and generates logical answers that include the most relevant legal provisions and explanations. It then sends the generated answers to the users.

[0044] Terminal-side embodiment

[0045] The terminal provides a user interface for users to input queries. This user interface is implemented, for example, as a web browser or mobile application, allowing users to easily enter questions.

[0046] The terminal sends queries entered by the user to the server. These queries are sent using HTTP requests or other communication protocols. The terminal is also responsible for receiving the response from the server and displaying it to the user.

[0047] User-side embodiment

[0048] The user first enters a question about a specific law through the terminal's user interface. This query might be something like, "Please tell me how to convene a shareholders' meeting." The entered query is then sent to the server via the terminal.

[0049] The user checks the response sent from the server via their terminal. For example, a logical response such as, "In accordance with Article AA of the Companies Act, the following procedures must be followed to convene a shareholders' meeting," is displayed. In this way, users can quickly obtain accurate information regarding laws and regulations.

[0050] Specific example

[0051] For example, consider a case where a user enters the following query: "Please tell me about the procedures for establishing a company."

[0052] 1. The user enters the query using the terminal's user interface.

[0053] 2. The query is sent from the terminal to the server.

[0054] 3. The server receives the query and performs analysis.

[0055] 4. Use a generative artificial intelligence model to search the knowledge database and generate the optimal answer.

[0056] 5. The server sends the generated response to the terminal.

[0057] 6. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[0058] The above describes the embodiments of the present invention. This allows companies to efficiently handle complex legal compliance requirements and reduce significant effort and costs.

[0059] The following describes the processing flow.

[0060] Server-side processing steps

[0061] Step 1:

[0062] The server collects legal data. Specifically, it gathers legal texts, case studies, and expert opinions from the internet and legal databases, and inputs them into the system.

[0063] Step 2:

[0064] The server formats the collected legal data. This includes text normalization (e.g., standardizing uppercase and lowercase letters, removing unnecessary spaces, etc.) and data structuring (tagging and standardizing formatting).

[0065] Step 3:

[0066] The server trains a generative AI model using formatted legal data. The AI ​​model uses multi-layer neural networks and other methods to extract semantic information from the legal data and learn from it.

[0067] Step 4:

[0068] The server uses a trained generative AI model to build a knowledge database. Knowledge extraction primarily involves semantic analysis of text to create a foundation for generating legally compliant responses.

[0069] Step 5:

[0070] The server receives queries from users. These queries are received as text sent via HTTP requests, etc.

[0071] Step 6:

[0072] The server parses the received query. Based on the query's content, it searches the knowledge database to identify relevant legal provisions and explanations.

[0073] Step 7:

[0074] The server generates logical answers from relevant legal information. Using an AI model, it provides the most appropriate information in a sound and logical manner to user queries.

[0075] Step 8:

[0076] The server sends the generated response to the terminal. This also uses protocols such as HTTP responses to return data to the terminal.

[0077] Terminal-side processing steps

[0078] Step 1:

[0079] The terminal accepts query input from the user. This process takes place through the user interface of a web browser or mobile app.

[0080] Step 2:

[0081] The terminal sends user-entered queries to the server. Specifically, it sends data to the server using HTTP requests and API calls.

[0082] Step 3:

[0083] The terminal receives a response from the server. It analyzes the response data sent from the server and displays it appropriately in the user interface.

[0084] Step 4:

[0085] The terminal displays the response received from the server to the user. Text formatting and GUI elements are used to ensure the response is displayed clearly and understandably.

[0086] User-side processing steps

[0087] Step 1:

[0088] Users enter queries related to laws and regulations through the terminal's user interface. For example, they might enter a specific question such as, "Please tell me how to convene a shareholders' meeting."

[0089] Step 2:

[0090] The user sends the entered query to the server via their terminal. This ensures that the user's question reaches the server.

[0091] Step 3:

[0092] The user checks the response from the server displayed on their terminal. They view explanations of laws and procedures displayed on their terminal and use them in their actual work as needed. For example, they check the procedures for company establishment and follow those steps.

[0093] (Example 1)

[0094] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0095] In corporate legal affairs, it is crucial to quickly and accurately understand and implement complex regulatory provisions and procedures. However, legal documents themselves are vast in volume, making it difficult to extract necessary information and make appropriate judgments. Furthermore, considering expert opinions and past precedents is time-consuming and costly. Moreover, the scope of law is broad, making it difficult for the average corporate employee to deeply understand the regulations. There is a need for methods to solve these challenges and streamline corporate legal affairs.

[0096] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0097] In this invention, the server includes means for learning regulatory data using a generative artificial intelligence model and building a knowledge base; means for receiving regulatory questions from users; means for referencing appropriate regulatory statements from the knowledge base based on the questions and generating logical answers; means for providing the generated answers to users; means for collecting regulatory data and formatting it using natural language processing technology; means for generating answers from regulatory data using a generative AI model; and means for transmitting the generated answers to a terminal using a communication protocol. This makes it possible to quickly and accurately extract complex information regarding laws and regulations and provide users with appropriate answers.

[0098] A "generative artificial intelligence model" is an algorithm that learns from existing data and generates new information or text based on that information.

[0099] "Regulatory data" refers to legal information, including laws, regulations, precedents, and expert opinions.

[0100] A "knowledge base" is a database that systematically organizes and stores knowledge related to a specific field.

[0101] A "user" is someone who uses the system to input questions and receive answers.

[0102] A "question" is a specific information request that a user makes to a knowledge base.

[0103] A "regulatory document" is an official document or article based on laws and regulations.

[0104] A "logical answer" is a response that includes appropriate information and legal explanations in response to a question, and is logically consistent.

[0105] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and generate human language.

[0106] A "communication protocol" is a set of prescribed procedures or methods for sending and receiving data over a network.

[0107] A "terminal" is a device used by users to access a system, input questions, and receive answers.

[0108] This invention relates to a system that efficiently processes regulatory data using a generative AI model to support companies' legal compliance. This system operates through the cooperation of the server, terminal, and user.

[0109] Server-side embodiment

[0110] The server first collects and formats regulatory data. Specifically, it automatically collects data such as legal documents, case precedents, and expert opinions using web scraping and APIs. This task utilizes Python libraries such as BeautifulSoup and Scrapy.

[0111] The collected data is formatted using natural language processing (NLP) techniques. Python libraries such as spaCy and NLTK are used for text analysis and information extraction. As a result, the data is structured and stored in a knowledge base.

[0112] Next, the server trains the data using a generative AI model. Specifically, generative AI models such as GPT-4 are used. The dataset is input into the model, and it is trained using a multi-layer neural network. Data preprocessing and tokenization are also performed during this process.

[0113] After training, the server receives queries from users and performs analysis. Queries are sent to the server as HTTP requests, and natural language processing techniques are used for analysis. Based on the analysis results, the server searches a knowledge base and generates the most appropriate legal text or answer.

[0114] Terminal-side embodiment

[0115] The terminal provides an interface for the user to enter queries. This interface is implemented as a web browser or mobile application. The user uses the terminal to enter specific questions, which are then sent to the server as HTTP requests.

[0116] The terminal receives responses from the server and displays them in the user interface. This allows the user to obtain information quickly and accurately.

[0117] User-side embodiment

[0118] The user first enters a specific question through the terminal's user interface. For example, a question such as, "Please tell me how to convene a shareholders' meeting." The entered query is sent to the server via the terminal.

[0119] The user checks the response sent from the server on their device. For example, a logical response such as, "In accordance with Article AA of the Companies Act, the following procedures must be followed to convene a shareholders' meeting," will be displayed.

[0120] Specific example

[0121] For example, consider a case where a user enters the query, "Please tell me about the procedures for establishing a company."

[0122] 1. The user enters the query using the terminal's user interface.

[0123] 2. The query is sent from the terminal to the server.

[0124] 3. The server receives the query and performs analysis.

[0125] 4. Use a generative artificial intelligence model (e.g., GPT-4) to search the knowledge base and generate the optimal answer.

[0126] 5. The server sends the generated response to the terminal.

[0127] 6. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[0128] Thus, the present invention is a system that provides complex regulatory information quickly and accurately, thereby streamlining corporate legal compliance.

[0129] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0130] Step 1: Collection and formatting of regulatory data

[0131] The server automatically collects data such as legal documents, case precedents, and expert opinions using web scraping and APIs. Specifically, it uses Python libraries such as BeautifulSoup and Scrapy. The collected data is analyzed using natural language processing techniques with Python libraries such as spaCy and NLTK to extract important information. This generates structured data, which is then stored in a knowledge base.

[0132] Input: Legal documents, case law, expert opinions

[0133] Output: Structured regulatory data

[0134] Step 2: Training the Generative AI Model

[0135] The server trains a generative AI model (e.g., GPT-4) using regulatory data stored in its knowledge base. Specifically, it inputs the dataset into a multi-layer neural network model and trains it through text generation tasks. This process also includes data preprocessing and tokenization. This enables the model to generate appropriate answers to questions.

[0136] Input: Structured regulatory data

[0137] Output: Trained generative AI model

[0138] Step 3: Receive and parse queries

[0139] The terminal sends user queries to the server in the form of HTTP requests. The server analyzes the received queries using natural language processing techniques to extract topics and related keywords. This identifies the intent of the query.

[0140] Input: Query from user

[0141] Output: Analyzed query information (topics and keywords)

[0142] Step 4: Knowledge base search and answer generation

[0143] Based on the analyzed query information, the server searches its knowledge base to identify relevant legal texts and precedents. Next, the server uses a generative AI model to generate a logical response. Specifically, it creates an answer tailored to the user's query based on the identified legal texts and precedents.

[0144] Input: Parsed query information

[0145] Output: Generated logical answer

[0146] Step 5: Submit and view your response

[0147] The server sends the generated response to the terminal as an HTTP response. The terminal displays the received response in the user interface. This allows the user to immediately confirm the appropriate answer to the question.

[0148] Input: Generated logical answer

[0149] Output: Answer displayed in the user interface

[0150] (Application Example 1)

[0151] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0152] In industries such as electronic payment services, legal compliance is extremely important. However, laws are complex, and keeping up-to-date requires considerable effort and time. This makes it difficult for legal departments and employees to respond quickly and accurately to legal-related questions they face on a daily basis. To solve this problem, there is a need for a system that efficiently learns legal data and provides appropriate answers to user questions immediately.

[0153] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0154] This invention includes a server comprising means for learning legal data using a generative artificial intelligence model and constructing a knowledge database, means for receiving legal queries from users, means for referencing appropriate legal provisions from the knowledge database based on the queries and generating logical answers, means for providing the generated answers to users, and means for immediately answering legal-related questions in electronic payment services using an application installed on a smartphone. This enables electronic payment service providers to comply with laws and regulations quickly and accurately and reduce legal risks.

[0155] A "generative artificial intelligence model" is a type of artificial intelligence that has the ability to learn from large amounts of data and generate new information or answers based on that data.

[0156] "Legal data" refers to a dataset containing legal information and norms, including legal texts, case studies, and expert opinions.

[0157] A "knowledge database" is a database that organizes and stores information related to a specific area of ​​knowledge, making it quickly accessible.

[0158] A "query" is a question or request that a user enters in order to obtain specific information or an answer.

[0159] A "logical answer" is an accurate answer that is logically consistent and based on evidence.

[0160] "Means of providing information to users" refers to all methods and devices used to transmit information generated by the system to end users.

[0161] A "smartphone" is a small, portable information terminal that has information processing and communication capabilities.

[0162] "Electronic payment services" refer to all services that allow users to make payments using the internet or mobile devices.

[0163]

[0164] In order to implement this invention, the following system configuration and processing are necessary.

[0165] Server-side embodiment

[0166] The server first collects and formats legal data. This legal data includes legal text, case examples, and expert opinions. This data is used to train a generative AI model and build a knowledge database. Next, the server receives legal queries from users. These queries are sent in the form of specific questions from the user. Based on the query, the server searches the knowledge database, refers to the appropriate legal provisions, and generates a logical answer. The generated answer is provided to the user through an application installed on their smartphone.

[0167] Terminal-side embodiment

[0168] The terminal receives queries from the user through a user interface. This user interface is implemented, for example, as a web browser or mobile application, allowing users to easily input questions. The terminal sends the entered queries to the server. Queries are sent using communication protocols such as HTTP requests. The terminal also receives the response from the server and displays it to the user.

[0169] User-side embodiment

[0170] The user first enters a question about specific laws and regulations through the terminal's user interface. This query might be something like, "What are the legal requirements for KYC procedures?" The entered query is sent to the server via the terminal. The user then checks the answer sent from the server via the terminal, resulting in the rapid provision of accurate information about the laws and regulations.

[0171] Data processing and calculation

[0172] The following processes are performed on the server side:

[0173] Hardware: Servers and smartphones

[0174] Software: Python, Flask, HTTP communication library

[0175] First, a knowledge database is constructed by training an AI model with collected legal data. This knowledge database is designed to store the latest and most appropriate information regarding laws and regulations. Next, semantic information is extracted from the knowledge database to generate answers to user queries. Logical answers are generated through data retrieval and semantic information extraction based on the queries.

[0176] Specific example

[0177] For example, if a user enters the query "What are the legal requirements for KYC procedures for electronic payment services?" into an application when adding a new feature to an electronic payment platform for a fintech service, that query will be sent directly to the AI ​​model with a prompt like the following.

[0178] Example of a prompt:

[0179] "Could you please explain the legal requirements in Japan regarding KYC procedures?"

[0180] The above system enables electronic payment service providers to comply with laws and regulations quickly and accurately, thereby reducing legal risks.

[0181] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0182] Step 1:

[0183] The user enters specific questions (queries) regarding laws and regulations through the terminal's user interface. These queries might take the form of, for example, "What are the legal requirements for KYC procedures for electronic payment services?" The entered queries are collected by the terminal.

[0184] Step 2:

[0185] The terminal sends the collected queries to the server using communication protocols such as HTTP requests. During this process, the queries are formatted appropriately and sent in a way that makes them easy for the server to parse. The input is the user's query, and the output is the request containing that query.

[0186] Step 3:

[0187] The server parses the received queries. Natural language processing techniques are used for the analysis to extract the intent and keywords of the queries. Specifically, it identifies important information such as KYC procedures and legal requirements from the queries. The input is the query from the terminal, and the output is structured data with the analysis results.

[0188] Step 4:

[0189] The server searches the knowledge database based on the analysis results. The knowledge database contains legal data, and a generative AI model is used to select the appropriate legal provisions and explanations. The input is the analysis results, and the output is the optimal legal information. Keyword matching and prediction results from machine learning models are used for the search.

[0190] Step 5:

[0191] Using a generative AI model, logical answers are generated based on information extracted from a knowledge database. These answers include direct and specific explanations for the user's query. The input is legal information from the knowledge database, and the output is the generated answer. Text generation algorithms and transformer models are used in the generation process.

[0192] Step 6:

[0193] The server then sends the generated response back to the terminal. An HTTP response is used for transmission, ensuring the information is sent to the terminal in the appropriate format. The input is the generated response, and the output is the response containing the response.

[0194] Step 7:

[0195] The terminal displays the received response to the user through a user interface. Based on this response, the user can quickly obtain specific information regarding the law. The input is the response from the server, and the output is the information displayed on the user's screen.

[0196] Through the steps described above, users can quickly and accurately obtain specific information for compliance with laws and regulations regarding electronic payment services.

[0197] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0198] This invention combines a system that uses a generative artificial intelligence model to learn legal data and build a knowledge database with an emotion engine that recognizes user emotions, thereby enabling the provision of legal information tailored to the user's emotions. The following describes how the server, terminal, and user interact to realize the system.

[0199] Server-side embodiment

[0200] The server first collects and formats legal data. This legal data includes legal text, case studies, and expert opinions, which are then formatted appropriately. The formatted data is then aggregated into a knowledge database using a generative AI model.

[0201] Next, the server trains a generative artificial intelligence model. This training is performed using collected legal data, and it is a process in which the AI ​​model extracts and learns semantic information from the legal data. The trained model has the ability to generate appropriate legal information from the knowledge database.

[0202] The server also receives and analyzes queries from users. These queries relate to company law and are submitted by users in the form of specific questions. Based on the queries, the server searches its knowledge database, identifies relevant legal provisions and explanations, and generates answers. Furthermore, it uses an emotion engine to analyze the user's emotions and adjust the tone and expression of the answers accordingly.

[0203] Terminal-side embodiment

[0204] The terminal provides a user interface for receiving query input from the user. This user interface is implemented, for example, as a web browser or mobile application, allowing the user to easily input questions.

[0205] The terminal sends queries entered by the user to the server. Queries are sent using HTTP requests or other communication protocols. The terminal also receives responses from the server and displays them to the user. The displayed responses are generated by the server and adjusted by the sentiment engine.

[0206] User-side embodiment

[0207] The user first enters a question about a specific law through the terminal's user interface. This query might be something like, "Please tell me how to convene a shareholders' meeting." The entered query is then sent to the server via the terminal.

[0208] The server's response is analyzed by an emotion engine and then displayed on the terminal with adjusted tone and expression to match the user's emotions. The user then reviews the displayed legal explanations and procedures and uses them in their actual work.

[0209] Specific example

[0210] Consider a scenario where a user enters the following query: "Please tell me about the procedures for establishing a company."

[0211] 1. The user enters the query using the terminal's user interface.

[0212] 2. The query is sent from the terminal to the server.

[0213] 3. The server receives the query and performs analysis.

[0214] 4. Use a generative artificial intelligence model to search the knowledge database and generate the optimal answer.

[0215] 5. The emotion engine analyzes the user's emotions at the time of input and adjusts the tone and expression of the response as needed. For example, if it determines that the user is nervous, it will generate a response in a calmer and more friendly tone.

[0216] 6. The server sends the generated response to the terminal.

[0217] 7. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[0218] This invention enables companies to respond quickly and appropriately to complex legal issues, as well as to provide more user-friendly and easily understandable information that resonates with users' emotions.

[0219] The following describes the processing flow.

[0220] Server-side processing steps

[0221] Step 1:

[0222] The server collects legal data. Specifically, it obtains legal texts, case studies, and expert opinions from the internet and legal databases, and inputs them into the system.

[0223] Step 2:

[0224] The server formats the collected legal data. This includes text normalization (unifying uppercase and lowercase letters, removing unnecessary spaces, etc.) and data structuring (converting to a specific format).

[0225] Step 3:

[0226] The server trains a generative AI model using formatted legal data. The AI ​​model extracts semantic information from the legal data and learns how to generate text.

[0227] Step 4:

[0228] The server uses a trained generative AI model to build a knowledge database. This knowledge database neatly stores various information related to company law.

[0229] Step 5:

[0230] The server receives queries from users. Queries are sent via HTTP requests or similar methods and parsed by the server.

[0231] Step 6:

[0232] The server searches the knowledge database based on the received query, identifying relevant legal provisions and case examples.

[0233] Step 7:

[0234] The server uses an emotion engine to analyze the user's emotions. It recognizes the user's emotions (e.g., tension, excitement, questioning, etc.) from the context and expression of the query.

[0235] Step 8:

[0236] The server adjusts the tone and expression of its response based on the emotion recognition results to generate the most appropriate response. For example, if the user is nervous, it will use a calm and encouraging tone.

[0237] Step 9:

[0238] The server sends the generated response to the terminal. It returns data to the terminal using protocols such as HTTP responses.

[0239] Terminal-side processing steps

[0240] Step 1:

[0241] The terminal accepts query input from the user. Input is received through a user interface (e.g., a web browser or mobile app).

[0242] Step 2:

[0243] The terminal sends the query entered by the user to the server. Data is transmitted via HTTP requests and API calls.

[0244] Step 3:

[0245] The terminal receives a response from the server. It analyzes the response data sent from the server and prepares it for display on the user interface.

[0246] Step 4:

[0247] The terminal displays the user the response received from the server. The response is presented in an easy-to-read format using text formatting and GUI elements.

[0248] User-side processing steps

[0249] Step 1:

[0250] The user enters a query regarding the law through the terminal's user interface. For example, they might enter the question, "Please tell me how to convene a shareholders' meeting."

[0251] Step 2:

[0252] The user operates a terminal to send a query to the server. This sends the user's question to the system.

[0253] Step 3:

[0254] The user checks the response from the server displayed on their terminal. They view the legal explanations and procedural guides displayed on their terminal and use them in their actual work as needed. For example, they might check the specific procedures and required documents for company establishment.

[0255] Through the above processing flow, the system provides quick and appropriate answers to user questions and enables flexible responses that respond to the user's emotions.

[0256] (Example 2)

[0257] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0258] Conventional legal search systems can provide appropriate legal information when users ask questions about laws, but the information provided is often rigid and not user-friendly. Furthermore, they fail to provide information that takes into account the user's feelings and circumstances, which can cause users to feel anxious or stressed. This invention aims to provide a method for providing legal information that also takes into account the user's feelings.

[0259] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0260] In this invention, the server includes means for learning legal data using a generative artificial intelligence model and constructing a knowledge database; means for receiving legal queries from users; means for referencing appropriate legal provisions and related information from the knowledge database based on the queries and generating logical answers; means for sentiment analysis for adjusting the generated answers according to the user's emotions; and means for providing the adjusted answers to the user. This makes it possible to provide legal information in a format that takes the user's emotions into consideration.

[0261] A "generative artificial intelligence model" is an artificial intelligence system that extracts semantic information from various forms of data based on given training data and performs appropriate responses and inferences.

[0262] "Legal data" refers to information on norms and legal guidelines, including legal texts, case studies, and expert opinions.

[0263] A "knowledge database" is a database that aggregates legal data learned by a generative AI model and stores it in a searchable format.

[0264] "Emotion analysis methods" are technologies that analyze user input and behavior to identify their emotional state.

[0265] An "emotion engine" is a part of a system that analyzes the user's emotions and adjusts the output accordingly.

[0266] A "query" is a question or request that a user enters into a system.

[0267] A "user interface" is the interface through which a user interacts with a system, and includes web browsers and mobile applications.

[0268] A "logical response" is a response aimed at providing consistent and logically correct information in response to a user's query.

[0269] This invention is a system that provides users with appropriate legal information by utilizing a generative artificial intelligence model based on data related to laws and regulations. This system primarily operates through the cooperation of three entities: a server, a terminal, and a user.

[0270] Server-side embodiment

[0271] The server first collects and formats legal data. This data is gathered from the internet and dedicated databases. The collected data includes legal text, case examples, and expert opinions. The collected data is then formatted by normalizing the text and removing unnecessary tags and formatting.

[0272] Next, the server trains a generative AI model using the formatted data. Training is performed using libraries such as Python's TensorFlow and PyTorch. During the training process, the AI ​​model extracts semantic information from legal data and acquires the ability to interpret laws and infer related information. The trained model is then stored in a knowledge database.

[0273] When a user submits a query, the server parses it. Natural language processing (NLP) techniques are used for query parsing. Libraries such as spaCy and NLTK are used in this process. Based on the parsed query, the server searches a knowledge database. It identifies relevant legal provisions and explanations from the search results and generates an answer using a generative AI model. Furthermore, it analyzes the user's emotions using an emotion engine and adjusts the tone and expression of the answer.

[0274] Terminal-side embodiment

[0275] The terminal provides a user interface. This user interface is implemented as a web browser or mobile application, allowing users to easily input questions. The terminal sends the user-entered query to the server. An HTTP POST request is used for transmission. Once the server returns a response, it is displayed in the user interface. The displayed response is then refined by an emotion engine.

[0276] User-side embodiment

[0277] The user first inputs a question regarding a specific law through the user interface of the terminal. For example, the user inputs a specific question such as "Please tell me about the method of convoking a shareholders' meeting." The input query is sent to the server via the terminal. The response from the server is analyzed by the emotion engine and displayed on the terminal as content with the tone and expression adjusted according to the user's emotion. The user checks the displayed explanation and procedure method regarding the law and applies it to actual business.

[0278] Specific example

[0279] When the user inputs a query "Please tell me about the company establishment procedure", the specific processing flow is as follows:

[0280] 1. The user uses the user interface of the terminal to input a query "Please tell me about the company establishment procedure".

[0281] 2. The terminal sends the query to the server (HTTP POST request).

[0282] 3. The server receives the query and analyzes it with natural language processing. It generates an answer from the knowledge database using a generative AI model.

[0283] 4. The emotion engine analyzes the emotion at the time of the user's input and adjusts the tone and expression of the answer. For example, when tension is felt, it displays "Please rest assured. The company establishment procedure is as follows...".

[0284] 5. The server sends the generated answer to the terminal.

[0285] 6. The terminal displays the answer on the user interface.

[0286] 7. The user checks the displayed answer and applies it as "The procedures required for company establishment are as follows:...".

[0287] This program enables companies to respond quickly and appropriately to complex legal issues, and also allows them to provide information that is sensitive to the user's feelings.

[0288] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0289] Step 1: Collection and formatting of legal data

[0290] The server collects legal texts, case studies, and expert opinions from the internet and dedicated databases. The collected data undergoes formatting processes such as text normalization and removal of unnecessary tags and formatting. Examples include the removal of HTML tags and classification by specific keywords.

[0291] Input: Legal text, case examples, expert opinions

[0292] Output: Normalized and formatted legal data

[0293] Step 2: Training the generative artificial intelligence model

[0294] The server uses formatted legal data to train a generative AI model. TensorFlow and PyTorch are used for this training. The model learns patterns and semantic information from the data, developing the ability to interpret laws and infer related information.

[0295] Input: Formatted legal data

[0296] Output: Trained Generative AI Model

[0297] Step 3: Receiving User Queries

[0298] The terminal accepts legal queries from users through its user interface. Users might enter questions such as, "Please tell me how to convene a shareholders' meeting."

[0299] Input: User's query

[0300] Output: Unprocessed query sent to the server

[0301] Step 4: Query analysis

[0302] The server receives the query sent from the terminal and analyzes it using natural language processing (NLP) techniques. Libraries such as spaCy and NLTK are used for this analysis. Understand the content of the query and convert it into appropriate search keys.

[0303] Input: Unprocessed query

[0304] Output: Analyzed query

[0305] Step 5: Search in the knowledge database

[0306] Based on the analyzed query, the server searches the knowledge database. Use a generative AI model to identify relevant legal articles and explanations and generate an optimal answer.

[0307] Input: Analyzed query

[0308] Output: Initial answer

[0309] Step 6: Sentiment analysis and tone adjustment

[0310] The server analyzes the sentiment from the user's query using a sentiment engine. If the user is nervous, adjust the tone of the answer to be calming. For example, if the sentiment analysis engine determines "uneasy" from the user's input, add a phrase like "Please rest assured."

[0311] Input: Initial answer

[0312] Output: Adjusted answer

[0313] Step 7: Providing responses to users

[0314] The server sends the adjusted response to the terminal. The terminal displays the response to the user through the user interface. This allows the user to check specific legal information and procedures.

[0315] Input: Adjusted answer

[0316] Output: Final answer displayed on the user interface

[0317] As described above, this system involves collaboration between the server, terminal, and user to provide legal information that takes user emotions into consideration.

[0318] (Application Example 2)

[0319] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0320] Traditionally, legal information systems have been able to generate logical answers to user queries, but they have not adequately considered user emotions when providing information. This may lead to decreased user satisfaction in emergencies or stressful situations. Furthermore, to alleviate user anxiety, the method of providing legal information needs to be improved. In particular, for urgent legal information related to safety measures, it is necessary to provide information in an appropriate tone that responds to user emotions.

[0321] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for learning legal data using a generative artificial intelligence model and constructing a knowledge database, means for providing a user interface equipped with an emotion engine that recognizes the user's emotions, means for receiving legal queries from the user and analyzing emotion data, means for referring to appropriate legal articles from the knowledge database based on the query and generating a logical response in a tone that matches the user's emotions, and means for providing the generated response to the user's terminal. This makes it possible to provide legal information that takes the user's emotions into consideration, enabling the provision of information that is familiar and reassuring to the user even in emergencies or stressful situations. Furthermore, this improves user satisfaction and achieves prompt and appropriate legal compliance.

[0322] A "generative artificial intelligence model" is an artificial intelligence algorithm that automatically learns from collected data and performs a specific task.

[0323] "Legal data" refers to data that includes information related to the law, such as laws, regulations, precedents, and expert opinions.

[0324] A "knowledge database" is a database that organizes knowledge extracted from legal data and formats it in a way that makes it easy to search and refer to.

[0325] An "emotion engine" is a software component that analyzes the user's emotional state and adjusts the tone and content of its responses based on that analysis.

[0326] A "user interface" is an interface that includes input and output means for a user to interact with a system.

[0327] A "query" is a question or request that a user enters into a system to obtain specific information.

[0328] "Emotional data" refers to data that indicates the emotional state of a user, extracted from their facial expressions, voice, and text.

[0329] A "logical answer" is an accurate and appropriate answer that is logically constructed based on legal data.

[0330] "Tone" refers to the way a generated response is expressed and spoken, and it is an element that is adjusted according to the user's emotions.

[0331] This invention relates to a system that uses a generative artificial intelligence model to learn from legal data and provides sentiment-sensitive legal information in response to user queries. This system consists of three components: a server, a terminal, and a user.

[0332] Server-side embodiment

[0333] The server first collects and formats legal data. This legal data includes legal texts, case precedents, and expert opinions. The formatted data is then aggregated into a knowledge database using a generative AI model.

[0334] Next, the server trains a generative artificial intelligence model (e.g., GPT-4). Training is performed based on collected legal data, and it is a process in which the AI ​​model extracts semantic information from the legal data and learns from it. The trained model has the ability to generate appropriate legal information from the knowledge database.

[0335] The server also receives and analyzes queries from users. These queries include sentiment data (e.g., facial expressions, voice data). Based on the queries, the server searches a knowledge database to identify relevant legal provisions and explanations, and uses a sentiment engine to adjust the tone of the generated responses.

[0336] Terminal-side embodiment

[0337] The device provides a user interface for receiving query input from the user. This user interface is implemented as a smartphone application, allowing users to input questions via voice or text. It also utilizes the device's built-in camera and microphone for emotion recognition.

[0338] Queries and sentiment data are sent to the server using HTTP requests and other communication protocols. Responses from the server are received in real time and displayed on the user interface in an adjusted format.

[0339] User-side embodiment

[0340] The user first enters a question through the terminal's user interface. For example, they might enter, "What legal action should I take if a suspicious person enters my home?" This query is then sent to the server via the terminal.

[0341] The response from the server is adjusted by an emotion engine to match the user's emotions in terms of tone and expression, and then resent to the device. Specifically, if the user is feeling anxious, the response will be provided in a calm tone, such as, "Don't worry. Please follow these steps to ensure your safety."

[0342] Example of a prompt

[0343] If the user is feeling tense and anxious, generate a gentle response regarding how to deal with a suspicious person. Keep it simple and specific, including examples of responses that are easy for the user to understand.

[0344] This system allows users to quickly obtain appropriate legal information tailored to their emotions, even in emergencies. This leads to improved user satisfaction and enables prompt and appropriate legal responses.

[0345] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0346] Step 1:

[0347] The user enters queries through the device's user interface. For example, they might enter a query via voice or text, such as, "What legal action should I take if a suspicious person enters my home?" The system acquires query text as input, along with voice and facial expression data for sentiment analysis. This data is collected using the device's camera and microphone.

[0348] Step 2:

[0349] The terminal receives the entered query and sentiment data and sends it to the server. The input data includes the user's voice and facial expression data, as well as the query text. This data is sent to the server via an HTTP request.

[0350] Step 3:

[0351] On the server side, the emotion recognition module first analyzes the audio data and facial expression data to recognize the user's emotions. Specifically, it uses OpenCV and TensorFlow to extract emotional features from the audio and analyze the facial expressions acquired from the camera. The output of the emotion recognition is the emotion the user is feeling (e.g., anxiety, tension).

[0352] Step 4:

[0353] The server uses a generative AI model to search the knowledge database based on the query text and refer to appropriate legal information. Specifically, the generative AI model (e.g., GPT-4) extracts relevant information from legal data and generates a logical response. The query text and knowledge database are required as input, and the output is the generated legal information response.

[0354] Step 5:

[0355] The emotion engine receives the response generated by the generative AI model. The emotion engine adjusts the tone and expression of the response based on the user's emotions. For example, if the user is feeling anxious, the tone of the response will be calmed and adjusted to something like, "Don't worry. Please follow the steps below." The input requires the response from the generative AI model and the emotion recognition result, and the output is the adjusted response.

[0356] Step 6:

[0357] The server sends the adjusted response to the device. The received response is formatted in a way that is easy for the user to understand and sent back to the device as an HTTP response.

[0358] Step 7:

[0359] The terminal displays the reconciled response received from the server on the user interface. This response may be provided not only as text, but also as voice response or in other formats. This allows users to receive familiar and reassuring information in real time.

[0360] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0361] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0362] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0363] [Second Embodiment]

[0364] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0365] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0366] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0367] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0368] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0369] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0370] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0371] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0372] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0373] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0374] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0375] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0376] This invention is a system that utilizes generative AI to streamline the complex provisions and procedures of company law, thereby supporting companies' legal compliance. The following describes how the server, terminal, and user entities cooperate to implement the system.

[0377] Server-side embodiment

[0378] The server first collects and formats legal data. This legal data includes legal text, case studies, and expert opinions, which are then formatted appropriately. The formatted data is then aggregated into a knowledge database using a generative instinct artificial intelligence model.

[0379] Next, the server trains a generative artificial intelligence model. This training is performed by training a generative AI model, such as a multi-layer neural network, using the collected legal data. After training, the model will have the ability to extract semantic information from the knowledge database and generate answers to queries.

[0380] The server also receives and analyzes queries from users. These queries relate to company law and are submitted by users in the form of specific questions. Based on the queries, the server searches its knowledge database and generates logical answers that include the most relevant legal provisions and explanations. It then sends the generated answers to the users.

[0381] Terminal-side embodiment

[0382] The terminal provides a user interface for users to input queries. This user interface is implemented, for example, as a web browser or mobile application, allowing users to easily enter questions.

[0383] The terminal sends queries entered by the user to the server. These queries are sent using HTTP requests or other communication protocols. The terminal is also responsible for receiving the response from the server and displaying it to the user.

[0384] User-side embodiment

[0385] The user first enters a question about a specific law through the terminal's user interface. This query might be something like, "Please tell me how to convene a shareholders' meeting." The entered query is then sent to the server via the terminal.

[0386] The user checks the response sent from the server via their terminal. For example, a logical response such as, "In accordance with Article AA of the Companies Act, the following procedures must be followed to convene a shareholders' meeting," is displayed. In this way, users can quickly obtain accurate information regarding laws and regulations.

[0387] Specific example

[0388] For example, consider a case where a user enters the following query: "Please tell me about the procedures for establishing a company."

[0389] 1. The user enters the query using the terminal's user interface.

[0390] 2. The query is sent from the terminal to the server.

[0391] 3. The server receives the query and performs analysis.

[0392] 4. Use a generative artificial intelligence model to search the knowledge database and generate the optimal answer.

[0393] 5. The server sends the generated response to the terminal.

[0394] 6. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[0395] The above describes the embodiments of the present invention. This allows companies to efficiently handle complex legal compliance requirements and reduce significant effort and costs.

[0396] The following describes the processing flow.

[0397] Server-side processing steps

[0398] Step 1:

[0399] The server collects legal data. Specifically, it gathers legal texts, case studies, and expert opinions from the internet and legal databases, and inputs them into the system.

[0400] Step 2:

[0401] The server formats the collected legal data. This includes text normalization (e.g., standardizing uppercase and lowercase letters, removing unnecessary spaces, etc.) and data structuring (tagging and standardizing formatting).

[0402] Step 3:

[0403] The server trains a generative AI model using formatted legal data. The AI ​​model uses multi-layer neural networks and other methods to extract semantic information from the legal data and learn from it.

[0404] Step 4:

[0405] The server uses a trained generative AI model to build a knowledge database. Knowledge extraction primarily involves semantic analysis of text to create a foundation for generating legally compliant responses.

[0406] Step 5:

[0407] The server receives queries from users. These queries are received as text sent via HTTP requests, etc.

[0408] Step 6:

[0409] The server parses the received query. Based on the query's content, it searches the knowledge database to identify relevant legal provisions and explanations.

[0410] Step 7:

[0411] The server generates logical answers from relevant legal information. Using an AI model, it provides the most appropriate information in a sound and logical manner to user queries.

[0412] Step 8:

[0413] The server sends the generated response to the terminal. This also uses protocols such as HTTP responses to return data to the terminal.

[0414] Terminal-side processing steps

[0415] Step 1:

[0416] The terminal accepts query input from the user. This process takes place through the user interface of a web browser or mobile app.

[0417] Step 2:

[0418] The terminal sends user-entered queries to the server. Specifically, it sends data to the server using HTTP requests and API calls.

[0419] Step 3:

[0420] The terminal receives a response from the server. It analyzes the response data sent from the server and displays it appropriately in the user interface.

[0421] Step 4:

[0422] The terminal displays the response received from the server to the user. Text formatting and GUI elements are used to ensure the response is displayed clearly and understandably.

[0423] User-side processing steps

[0424] Step 1:

[0425] Users enter queries related to laws and regulations through the terminal's user interface. For example, they might enter a specific question such as, "Please tell me how to convene a shareholders' meeting."

[0426] Step 2:

[0427] The user sends the entered query to the server via their terminal. This ensures that the user's question reaches the server.

[0428] Step 3:

[0429] The user checks the response from the server displayed on their terminal. They view explanations of laws and procedures displayed on their terminal and use them in their actual work as needed. For example, they check the procedures for company establishment and follow those steps.

[0430] (Example 1)

[0431] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0432] In corporate legal affairs, it is crucial to quickly and accurately understand and implement complex regulatory provisions and procedures. However, legal documents themselves are vast in volume, making it difficult to extract necessary information and make appropriate judgments. Furthermore, considering expert opinions and past precedents is time-consuming and costly. Moreover, the scope of law is broad, making it difficult for the average corporate employee to deeply understand the regulations. There is a need for methods to solve these challenges and streamline corporate legal affairs.

[0433] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0434] In this invention, the server includes means for learning regulatory data using a generative artificial intelligence model and building a knowledge base; means for receiving regulatory questions from users; means for referencing appropriate regulatory statements from the knowledge base based on the questions and generating logical answers; means for providing the generated answers to users; means for collecting regulatory data and formatting it using natural language processing technology; means for generating answers from regulatory data using a generative AI model; and means for transmitting the generated answers to a terminal using a communication protocol. This makes it possible to quickly and accurately extract complex information regarding laws and regulations and provide users with appropriate answers.

[0435] A "generative artificial intelligence model" is an algorithm that learns from existing data and generates new information or text based on that information.

[0436] "Regulatory data" refers to legal information, including laws, regulations, precedents, and expert opinions.

[0437] A "knowledge base" is a database that systematically organizes and stores knowledge related to a specific field.

[0438] A "user" is someone who uses the system to input questions and receive answers.

[0439] A "question" is a specific information request that a user makes to a knowledge base.

[0440] A "regulatory document" is an official document or article based on laws and regulations.

[0441] A "logical answer" is a response that includes appropriate information and legal explanations in response to a question, and is logically consistent.

[0442] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and generate human language.

[0443] A "communication protocol" is a set of prescribed procedures or methods for sending and receiving data over a network.

[0444] A "terminal" is a device used by users to access a system, input questions, and receive answers.

[0445] This invention relates to a system that efficiently processes regulatory data using a generative AI model to support companies' legal compliance. This system operates through the cooperation of the server, terminal, and user.

[0446] Server-side embodiment

[0447] The server first collects and formats regulatory data. Specifically, it automatically collects data such as legal documents, case precedents, and expert opinions using web scraping and APIs. This task utilizes Python libraries such as BeautifulSoup and Scrapy.

[0448] The collected data is formatted using natural language processing (NLP) techniques. Python libraries such as spaCy and NLTK are used for text analysis and information extraction. As a result, the data is structured and stored in a knowledge base.

[0449] Next, the server trains the data using a generative AI model. Specifically, generative AI models such as GPT-4 are used. The dataset is input into the model, and it is trained using a multi-layer neural network. Data preprocessing and tokenization are also performed during this process.

[0450] After training, the server receives queries from users and performs analysis. Queries are sent to the server as HTTP requests, and natural language processing techniques are used for analysis. Based on the analysis results, the server searches a knowledge base and generates the most appropriate legal text or answer.

[0451] Terminal-side embodiment

[0452] The terminal provides an interface for the user to enter queries. This interface is implemented as a web browser or mobile application. The user uses the terminal to enter specific questions, which are then sent to the server as HTTP requests.

[0453] The terminal receives responses from the server and displays them in the user interface. This allows the user to obtain information quickly and accurately.

[0454] User-side embodiment

[0455] The user first enters a specific question through the terminal's user interface. For example, a question such as, "Please tell me how to convene a shareholders' meeting." The entered query is sent to the server via the terminal.

[0456] The user checks the response sent from the server on their device. For example, a logical response such as, "In accordance with Article AA of the Companies Act, the following procedures must be followed to convene a shareholders' meeting," will be displayed.

[0457] Specific example

[0458] For example, consider a case where a user enters the query, "Please tell me about the procedures for establishing a company."

[0459] 1. The user enters the query using the terminal's user interface.

[0460] 2. The query is sent from the terminal to the server.

[0461] 3. The server receives the query and performs analysis.

[0462] 4. Use a generative artificial intelligence model (e.g., GPT-4) to search the knowledge base and generate the optimal answer.

[0463] 5. The server sends the generated response to the terminal.

[0464] 6. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[0465] Thus, the present invention is a system that provides complex regulatory information quickly and accurately, thereby streamlining corporate legal compliance.

[0466] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0467] Step 1: Collection and formatting of regulatory data

[0468] The server automatically collects data such as legal documents, case precedents, and expert opinions using web scraping and APIs. Specifically, it uses Python libraries such as BeautifulSoup and Scrapy. The collected data is analyzed using natural language processing techniques with Python libraries such as spaCy and NLTK to extract important information. This generates structured data, which is then stored in a knowledge base.

[0469] Input: Legal documents, case law, expert opinions

[0470] Output: Structured regulatory data

[0471] Step 2: Training the Generative AI Model

[0472] The server trains a generative AI model (e.g., GPT-4) using regulatory data stored in its knowledge base. Specifically, it inputs the dataset into a multi-layer neural network model and trains it through text generation tasks. This process also includes data preprocessing and tokenization. This enables the model to generate appropriate answers to questions.

[0473] Input: Structured regulatory data

[0474] Output: Trained generative AI model

[0475] Step 3: Receive and parse queries

[0476] The terminal sends user queries to the server in the form of HTTP requests. The server analyzes the received queries using natural language processing techniques to extract topics and related keywords. This identifies the intent of the query.

[0477] Input: Query from user

[0478] Output: Analyzed query information (topics and keywords)

[0479] Step 4: Knowledge base search and answer generation

[0480] Based on the analyzed query information, the server searches its knowledge base to identify relevant legal texts and precedents. Next, the server uses a generative AI model to generate a logical response. Specifically, it creates an answer tailored to the user's query based on the identified legal texts and precedents.

[0481] Input: Parsed query information

[0482] Output: Generated logical answer

[0483] Step 5: Submit and view your response

[0484] The server sends the generated response to the terminal as an HTTP response. The terminal displays the received response in the user interface. This allows the user to immediately confirm the appropriate answer to the question.

[0485] Input: Generated logical answer

[0486] Output: Answer displayed in the user interface

[0487] (Application Example 1)

[0488] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0489] In industries such as electronic payment services, legal compliance is extremely important. However, laws are complex, and keeping up-to-date requires considerable effort and time. This makes it difficult for legal departments and employees to respond quickly and accurately to legal-related questions they face on a daily basis. To solve this problem, there is a need for a system that efficiently learns legal data and provides appropriate answers to user questions immediately.

[0490] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0491] This invention includes a server comprising means for learning legal data using a generative artificial intelligence model and constructing a knowledge database, means for receiving legal queries from users, means for referencing appropriate legal provisions from the knowledge database based on the queries and generating logical answers, means for providing the generated answers to users, and means for immediately answering legal-related questions in electronic payment services using an application installed on a smartphone. This enables electronic payment service providers to comply with laws and regulations quickly and accurately and reduce legal risks.

[0492] A "generative artificial intelligence model" is a type of artificial intelligence that has the ability to learn from large amounts of data and generate new information or answers based on that data.

[0493] "Legal data" refers to a dataset containing legal information and norms, including legal texts, case studies, and expert opinions.

[0494] A "knowledge database" is a database that organizes and stores information related to a specific area of ​​knowledge, making it quickly accessible.

[0495] A "query" is a question or request that a user enters in order to obtain specific information or an answer.

[0496] A "logical answer" is an accurate answer that is logically consistent and based on evidence.

[0497] "Means of providing information to users" refers to all methods and devices used to transmit information generated by the system to end users.

[0498] A "smartphone" is a small, portable information terminal that has information processing and communication capabilities.

[0499] "Electronic payment services" refer to all services that allow users to make payments using the internet or mobile devices.

[0500]

[0501] In order to implement this invention, the following system configuration and processing are necessary.

[0502] Server-side embodiment

[0503] The server first collects and formats legal data. This legal data includes legal text, case examples, and expert opinions. This data is used to train a generative AI model and build a knowledge database. Next, the server receives legal queries from users. These queries are sent in the form of specific questions from the user. Based on the query, the server searches the knowledge database, refers to the appropriate legal provisions, and generates a logical answer. The generated answer is provided to the user through an application installed on their smartphone.

[0504] Terminal-side embodiment

[0505] The terminal receives queries from the user through a user interface. This user interface is implemented, for example, as a web browser or mobile application, allowing users to easily input questions. The terminal sends the entered queries to the server. Queries are sent using communication protocols such as HTTP requests. The terminal also receives the response from the server and displays it to the user.

[0506] User-side embodiment

[0507] The user first enters a question about specific laws and regulations through the terminal's user interface. This query might be something like, "What are the legal requirements for KYC procedures?" The entered query is sent to the server via the terminal. The user then checks the answer sent from the server via the terminal, resulting in the rapid provision of accurate information about the laws and regulations.

[0508] Data processing and calculation

[0509] The following processes are performed on the server side:

[0510] Hardware: Servers and smartphones

[0511] Software: Python, Flask, HTTP communication library

[0512] First, a knowledge database is constructed by training an AI model with collected legal data. This knowledge database is designed to store the latest and most appropriate information regarding laws and regulations. Next, semantic information is extracted from the knowledge database to generate answers to user queries. Logical answers are generated through data retrieval and semantic information extraction based on the queries.

[0513] Specific example

[0514] For example, if a user enters the query "What are the legal requirements for KYC procedures for electronic payment services?" into an application when adding a new feature to an electronic payment platform for a fintech service, that query will be sent directly to the AI ​​model with a prompt like the following.

[0515] Example of a prompt:

[0516] "Could you please explain the legal requirements in Japan regarding KYC procedures?"

[0517] The above system enables electronic payment service providers to comply with laws and regulations quickly and accurately, thereby reducing legal risks.

[0518] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0519] Step 1:

[0520] The user enters specific questions (queries) regarding laws and regulations through the terminal's user interface. These queries might take the form of, for example, "What are the legal requirements for KYC procedures for electronic payment services?" The entered queries are collected by the terminal.

[0521] Step 2:

[0522] The terminal sends the collected queries to the server using communication protocols such as HTTP requests. During this process, the queries are formatted appropriately and sent in a way that makes them easy for the server to parse. The input is the user's query, and the output is the request containing that query.

[0523] Step 3:

[0524] The server parses the received queries. Natural language processing techniques are used for the analysis to extract the intent and keywords of the queries. Specifically, it identifies important information such as KYC procedures and legal requirements from the queries. The input is the query from the terminal, and the output is structured data with the analysis results.

[0525] Step 4:

[0526] The server searches the knowledge database based on the analysis results. The knowledge database contains legal data, and a generative AI model is used to select the appropriate legal provisions and explanations. The input is the analysis results, and the output is the optimal legal information. Keyword matching and prediction results from machine learning models are used for the search.

[0527] Step 5:

[0528] Using a generative AI model, logical answers are generated based on information extracted from a knowledge database. These answers include direct and specific explanations for the user's query. The input is legal information from the knowledge database, and the output is the generated answer. Text generation algorithms and transformer models are used in the generation process.

[0529] Step 6:

[0530] The server then sends the generated response back to the terminal. An HTTP response is used for transmission, ensuring the information is sent to the terminal in the appropriate format. The input is the generated response, and the output is the response containing the response.

[0531] Step 7:

[0532] The terminal displays the received response to the user through a user interface. Based on this response, the user can quickly obtain specific information regarding the law. The input is the response from the server, and the output is the information displayed on the user's screen.

[0533] Through the steps described above, users can quickly and accurately obtain specific information for compliance with laws and regulations regarding electronic payment services.

[0534] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0535] This invention combines a system that uses a generative artificial intelligence model to learn legal data and build a knowledge database with an emotion engine that recognizes user emotions, thereby enabling the provision of legal information tailored to the user's emotions. The following describes how the server, terminal, and user interact to realize the system.

[0536] Server-side embodiment

[0537] The server first collects and formats legal data. This legal data includes legal text, case studies, and expert opinions, which are then formatted appropriately. The formatted data is then aggregated into a knowledge database using a generative AI model.

[0538] Next, the server trains a generative artificial intelligence model. This training is performed using collected legal data, and it is a process in which the AI ​​model extracts and learns semantic information from the legal data. The trained model has the ability to generate appropriate legal information from the knowledge database.

[0539] The server also receives and analyzes queries from users. These queries relate to company law and are submitted by users in the form of specific questions. Based on the queries, the server searches its knowledge database, identifies relevant legal provisions and explanations, and generates answers. Furthermore, it uses an emotion engine to analyze the user's emotions and adjust the tone and expression of the answers accordingly.

[0540] Terminal-side embodiment

[0541] The terminal provides a user interface for receiving query input from the user. This user interface is implemented, for example, as a web browser or mobile application, allowing the user to easily input questions.

[0542] The terminal sends queries entered by the user to the server. Queries are sent using HTTP requests or other communication protocols. The terminal also receives responses from the server and displays them to the user. The displayed responses are generated by the server and adjusted by the sentiment engine.

[0543] User-side embodiment

[0544] The user first enters a question about a specific law through the terminal's user interface. This query might be something like, "Please tell me how to convene a shareholders' meeting." The entered query is then sent to the server via the terminal.

[0545] The server's response is analyzed by an emotion engine and then displayed on the terminal with adjusted tone and expression to match the user's emotions. The user then reviews the displayed legal explanations and procedures and uses them in their actual work.

[0546] Specific example

[0547] Consider a scenario where a user enters the following query: "Please tell me about the procedures for establishing a company."

[0548] 1. The user enters the query using the terminal's user interface.

[0549] 2. The query is sent from the terminal to the server.

[0550] 3. The server receives the query and performs analysis.

[0551] 4. Use a generative artificial intelligence model to search the knowledge database and generate the optimal answer.

[0552] 5. The emotion engine analyzes the user's emotions at the time of input and adjusts the tone and expression of the response as needed. For example, if it determines that the user is nervous, it will generate a response in a calmer and more friendly tone.

[0553] 6. The server sends the generated response to the terminal.

[0554] 7. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[0555] This invention enables companies to respond quickly and appropriately to complex legal issues, as well as to provide more user-friendly and easily understandable information that resonates with users' emotions.

[0556] The following describes the processing flow.

[0557] Server-side processing steps

[0558] Step 1:

[0559] The server collects legal data. Specifically, it obtains legal texts, case studies, and expert opinions from the internet and legal databases, and inputs them into the system.

[0560] Step 2:

[0561] The server formats the collected legal data. This includes text normalization (unifying uppercase and lowercase letters, removing unnecessary spaces, etc.) and data structuring (converting to a specific format).

[0562] Step 3:

[0563] The server trains a generative AI model using formatted legal data. The AI ​​model extracts semantic information from the legal data and learns how to generate text.

[0564] Step 4:

[0565] The server uses a trained generative AI model to build a knowledge database. This knowledge database neatly stores various information related to company law.

[0566] Step 5:

[0567] The server receives queries from users. Queries are sent via HTTP requests or similar methods and parsed by the server.

[0568] Step 6:

[0569] The server searches the knowledge database based on the received query, identifying relevant legal provisions and case examples.

[0570] Step 7:

[0571] The server uses an emotion engine to analyze the user's emotions. It recognizes the user's emotions (e.g., tension, excitement, questioning, etc.) from the context and expression of the query.

[0572] Step 8:

[0573] The server adjusts the tone and expression of its response based on the emotion recognition results to generate the most appropriate response. For example, if the user is nervous, it will use a calm and encouraging tone.

[0574] Step 9:

[0575] The server sends the generated response to the terminal. It returns data to the terminal using protocols such as HTTP responses.

[0576] Terminal-side processing steps

[0577] Step 1:

[0578] The terminal accepts query input from the user. Input is received through a user interface (e.g., a web browser or mobile app).

[0579] Step 2:

[0580] The terminal sends the query entered by the user to the server. Data is transmitted via HTTP requests and API calls.

[0581] Step 3:

[0582] The terminal receives a response from the server. It analyzes the response data sent from the server and prepares it for display on the user interface.

[0583] Step 4:

[0584] The terminal displays the user the response received from the server. The response is presented in an easy-to-read format using text formatting and GUI elements.

[0585] User-side processing steps

[0586] Step 1:

[0587] The user enters a query regarding the law through the terminal's user interface. For example, they might enter the question, "Please tell me how to convene a shareholders' meeting."

[0588] Step 2:

[0589] The user operates a terminal to send a query to the server. This sends the user's question to the system.

[0590] Step 3:

[0591] The user checks the response from the server displayed on their terminal. They view the legal explanations and procedural guides displayed on their terminal and use them in their actual work as needed. For example, they might check the specific procedures and required documents for company establishment.

[0592] Through the above processing flow, the system provides quick and appropriate answers to user questions and enables flexible responses that respond to the user's emotions.

[0593] (Example 2)

[0594] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0595] Conventional legal search systems can provide appropriate legal information when users ask questions about laws, but the information provided is often rigid and not user-friendly. Furthermore, they fail to provide information that takes into account the user's feelings and circumstances, which can cause users to feel anxious or stressed. This invention aims to provide a method for providing legal information that also takes into account the user's feelings.

[0596] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0597] In this invention, the server includes means for learning legal data using a generative artificial intelligence model and constructing a knowledge database; means for receiving legal queries from users; means for referencing appropriate legal provisions and related information from the knowledge database based on the queries and generating logical answers; means for sentiment analysis for adjusting the generated answers according to the user's emotions; and means for providing the adjusted answers to the user. This makes it possible to provide legal information in a format that takes the user's emotions into consideration.

[0598] A "generative artificial intelligence model" is an artificial intelligence system that extracts semantic information from various forms of data based on given training data and performs appropriate responses and inferences.

[0599] "Legal data" refers to information on norms and legal guidelines, including legal texts, case studies, and expert opinions.

[0600] A "knowledge database" is a database that aggregates legal data learned by a generative AI model and stores it in a searchable format.

[0601] "Emotion analysis methods" are technologies that analyze user input and behavior to identify their emotional state.

[0602] An "emotion engine" is a part of a system that analyzes the user's emotions and adjusts the output accordingly.

[0603] A "query" is a question or request that a user enters into a system.

[0604] A "user interface" is the interface through which a user interacts with a system, and includes web browsers and mobile applications.

[0605] A "logical response" is a response aimed at providing consistent and logically correct information in response to a user's query.

[0606] This invention is a system that provides users with appropriate legal information by utilizing a generative artificial intelligence model based on data related to laws and regulations. This system primarily operates through the cooperation of three entities: a server, a terminal, and a user.

[0607] Server-side embodiment

[0608] The server first collects and formats legal data. This data is gathered from the internet and dedicated databases. The collected data includes legal text, case examples, and expert opinions. The collected data is then formatted by normalizing the text and removing unnecessary tags and formatting.

[0609] Next, the server trains a generative AI model using the formatted data. Training is performed using libraries such as Python's TensorFlow and PyTorch. During the training process, the AI ​​model extracts semantic information from legal data and acquires the ability to interpret laws and infer related information. The trained model is then stored in a knowledge database.

[0610] When a user submits a query, the server parses it. Natural language processing (NLP) techniques are used for query parsing. Libraries such as spaCy and NLTK are used in this process. Based on the parsed query, the server searches a knowledge database. It identifies relevant legal provisions and explanations from the search results and generates an answer using a generative AI model. Furthermore, it analyzes the user's emotions using an emotion engine and adjusts the tone and expression of the answer.

[0611] Terminal-side embodiment

[0612] The terminal provides a user interface. This user interface is implemented as a web browser or mobile application, allowing users to easily input questions. The terminal sends the user-entered query to the server. An HTTP POST request is used for transmission. Once the server returns a response, it is displayed in the user interface. The displayed response is then refined by an emotion engine.

[0613] User-side embodiment

[0614] The user first enters a specific legal question through the terminal's user interface. For example, they might enter a question like, "Please tell me how to convene a shareholders' meeting." The entered query is sent to the server via the terminal. The server's response is analyzed by an emotion engine and displayed on the terminal with a tone and expression adjusted to match the user's emotions. The user then reviews the displayed explanation of the law and the procedures and uses it in their actual work.

[0615] Specific example

[0616] The specific processing flow when a user enters the query "Please tell me about the company incorporation procedure" is as follows:

[0617] 1. The user uses the terminal's user interface to enter the query, "Please tell me about the company incorporation procedure."

[0618] 2. The terminal sends a query to the server (HTTP POST request).

[0619] 3. The server receives the query and analyzes it using natural language processing. It then generates an answer from the knowledge database using a generative AI model.

[0620] 4. The emotion engine analyzes the user's emotions at the time of input and adjusts the tone and expression of the response. For example, if tension is detected, it will display, "Please rest assured. The company incorporation procedure is as follows..."

[0621] 5. The server sends the generated response to the terminal.

[0622] 6. The device displays the answer in the user interface.

[0623] 7. The user reviews the displayed answer and uses it to say, "The necessary procedures for company establishment are as follows:..."

[0624] This program enables companies to respond quickly and appropriately to complex legal issues, and also allows them to provide information that is sensitive to the user's feelings.

[0625] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0626] Step 1: Collection and formatting of legal data

[0627] The server collects legal texts, case studies, and expert opinions from the internet and dedicated databases. The collected data undergoes formatting processes such as text normalization and removal of unnecessary tags and formatting. Examples include the removal of HTML tags and classification by specific keywords.

[0628] Input: Legal text, case examples, expert opinions

[0629] Output: Normalized and formatted legal data

[0630] Step 2: Training the generative artificial intelligence model

[0631] The server uses formatted legal data to train a generative AI model. TensorFlow and PyTorch are used for this training. The model learns patterns and semantic information from the data, developing the ability to interpret laws and infer related information.

[0632] Input: Formatted legal data

[0633] Output: Trained Generative AI Model

[0634] Step 3: Receiving User Queries

[0635] The terminal accepts legal queries from users through its user interface. Users might enter questions such as, "Please tell me how to convene a shareholders' meeting."

[0636] Input: User query

[0637] Output: Unprocessed queries sent to the server

[0638] Step 4: Analyze the query

[0639] The server receives queries sent from terminals and analyzes them using natural language processing (NLP) techniques. Libraries such as spaCy and NLTK are used for this analysis. The server understands the content of the query and converts it into appropriate search keys.

[0640] Input: Unprocessed query

[0641] Output: Parsed query

[0642] Step 5: Search the knowledge database

[0643] The server searches the knowledge database based on the analyzed query. Using a generative AI model, it identifies relevant legal provisions and explanations and generates the optimal answer.

[0644] Input: Parsed query

[0645] Output: Initial draft answer

[0646] Step 6: Emotional Analysis and Tone Adjustment

[0647] The server uses an emotion engine to analyze the user's emotions from their queries. If the user is anxious, it adjusts the tone of its responses to be calmer. For example, if the emotion analysis engine determines from the user's input that they are "anxious," it might add the phrase "Please rest assured."

[0648] Input: Initial draft answer

[0649] Output: Adjusted answer

[0650] Step 7: Providing responses to users

[0651] The server sends the adjusted response to the terminal. The terminal displays the response to the user through the user interface. This allows the user to check specific legal information and procedures.

[0652] Input: Adjusted answer

[0653] Output: Final answer displayed on the user interface

[0654] As described above, this system involves collaboration between the server, terminal, and user to provide legal information that takes user emotions into consideration.

[0655] (Application Example 2)

[0656] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0657] Traditionally, legal information systems have been able to generate logical answers to user queries, but they have not adequately considered user emotions when providing information. This may lead to decreased user satisfaction in emergencies or stressful situations. Furthermore, to alleviate user anxiety, the method of providing legal information needs to be improved. In particular, for urgent legal information related to safety measures, it is necessary to provide information in an appropriate tone that responds to user emotions.

[0658] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for learning legal data using a generative artificial intelligence model and constructing a knowledge database, means for providing a user interface equipped with an emotion engine that recognizes the user's emotions, means for receiving legal queries from the user and analyzing emotion data, means for referring to appropriate legal articles from the knowledge database based on the query and generating a logical response in a tone that matches the user's emotions, and means for providing the generated response to the user's terminal. This makes it possible to provide legal information that takes the user's emotions into consideration, enabling the provision of information that is familiar and reassuring to the user even in emergencies or stressful situations. Furthermore, this improves user satisfaction and achieves prompt and appropriate legal compliance.

[0659] A "generative artificial intelligence model" is an artificial intelligence algorithm that automatically learns from collected data and performs a specific task.

[0660] "Legal data" refers to data that includes information related to the law, such as laws, regulations, precedents, and expert opinions.

[0661] A "knowledge database" is a database that organizes knowledge extracted from legal data and formats it in a way that makes it easy to search and refer to.

[0662] An "emotion engine" is a software component that analyzes the user's emotional state and adjusts the tone and content of its responses based on that analysis.

[0663] A "user interface" is an interface that includes input and output means for a user to interact with a system.

[0664] A "query" is a question or request that a user enters into a system to obtain specific information.

[0665] "Emotional data" refers to data that indicates the emotional state of a user, extracted from their facial expressions, voice, and text.

[0666] A "logical answer" is an accurate and appropriate answer that is logically constructed based on legal data.

[0667] "Tone" refers to the way a generated response is expressed and spoken, and it is an element that is adjusted according to the user's emotions.

[0668] This invention relates to a system that uses a generative artificial intelligence model to learn from legal data and provides sentiment-sensitive legal information in response to user queries. This system consists of three components: a server, a terminal, and a user.

[0669] Server-side embodiment

[0670] The server first collects and formats legal data. This legal data includes legal texts, case precedents, and expert opinions. The formatted data is then aggregated into a knowledge database using a generative AI model.

[0671] Next, the server trains a generative artificial intelligence model (e.g., GPT-4). Training is performed based on collected legal data, and it is a process in which the AI ​​model extracts semantic information from the legal data and learns from it. The trained model has the ability to generate appropriate legal information from the knowledge database.

[0672] The server also receives and analyzes queries from users. These queries include sentiment data (e.g., facial expressions, voice data). Based on the queries, the server searches a knowledge database to identify relevant legal provisions and explanations, and uses a sentiment engine to adjust the tone of the generated responses.

[0673] Terminal-side embodiment

[0674] The device provides a user interface for receiving query input from the user. This user interface is implemented as a smartphone application, allowing users to input questions via voice or text. It also utilizes the device's built-in camera and microphone for emotion recognition.

[0675] Queries and sentiment data are sent to the server using HTTP requests and other communication protocols. Responses from the server are received in real time and displayed on the user interface in an adjusted format.

[0676] User-side embodiment

[0677] The user first enters a question through the terminal's user interface. For example, they might enter, "What legal action should I take if a suspicious person enters my home?" This query is then sent to the server via the terminal.

[0678] The response from the server is adjusted by an emotion engine to match the user's emotions in terms of tone and expression, and then resent to the device. Specifically, if the user is feeling anxious, the response will be provided in a calm tone, such as, "Don't worry. Please follow these steps to ensure your safety."

[0679] Example of a prompt

[0680] If the user is feeling tense and anxious, generate a gentle response regarding how to deal with a suspicious person. Keep it simple and specific, including examples of responses that are easy for the user to understand.

[0681] This system allows users to quickly obtain appropriate legal information tailored to their emotions, even in emergencies. This leads to improved user satisfaction and enables prompt and appropriate legal responses.

[0682] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0683] Step 1:

[0684] The user enters queries through the device's user interface. For example, they might enter a query via voice or text, such as, "What legal action should I take if a suspicious person enters my home?" The system acquires query text as input, along with voice and facial expression data for sentiment analysis. This data is collected using the device's camera and microphone.

[0685] Step 2:

[0686] The terminal receives the entered query and sentiment data and sends it to the server. The input data includes the user's voice and facial expression data, as well as the query text. This data is sent to the server via an HTTP request.

[0687] Step 3:

[0688] On the server side, the emotion recognition module first analyzes the audio data and facial expression data to recognize the user's emotions. Specifically, it uses OpenCV and TensorFlow to extract emotional features from the audio and analyze the facial expressions acquired from the camera. The output of the emotion recognition is the emotion the user is feeling (e.g., anxiety, tension).

[0689] Step 4:

[0690] The server uses a generative AI model to search the knowledge database based on the query text and refer to appropriate legal information. Specifically, the generative AI model (e.g., GPT-4) extracts relevant information from legal data and generates a logical response. The query text and knowledge database are required as input, and the output is the generated legal information response.

[0691] Step 5:

[0692] The emotion engine receives the response generated by the generative AI model. The emotion engine adjusts the tone and expression of the response based on the user's emotions. For example, if the user is feeling anxious, the tone of the response will be calmed and adjusted to something like, "Don't worry. Please follow the steps below." The input requires the response from the generative AI model and the emotion recognition result, and the output is the adjusted response.

[0693] Step 6:

[0694] The server sends the adjusted response to the device. The received response is formatted in a way that is easy for the user to understand and sent back to the device as an HTTP response.

[0695] Step 7:

[0696] The terminal displays the reconciled response received from the server on the user interface. This response may be provided not only as text, but also as voice response or in other formats. This allows users to receive familiar and reassuring information in real time.

[0697] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0698] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0699] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0700] [Third Embodiment]

[0701] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0702] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0703] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0704] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0705] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0706] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0707] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0708] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0709] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0710] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0711] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0712] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0713] This invention is a system that utilizes generative AI to streamline the complex provisions and procedures of company law, thereby supporting companies' legal compliance. The following describes how the server, terminal, and user entities cooperate to implement the system.

[0714] Server-side embodiment

[0715] The server first collects and formats legal data. This legal data includes legal text, case studies, and expert opinions, which are then formatted appropriately. The formatted data is then aggregated into a knowledge database using a generative instinct artificial intelligence model.

[0716] Next, the server trains a generative artificial intelligence model. This training is performed by training a generative AI model, such as a multi-layer neural network, using the collected legal data. After training, the model will have the ability to extract semantic information from the knowledge database and generate answers to queries.

[0717] The server also receives and analyzes queries from users. These queries relate to company law and are submitted by users in the form of specific questions. Based on the queries, the server searches its knowledge database and generates logical answers that include the most relevant legal provisions and explanations. It then sends the generated answers to the users.

[0718] Terminal-side embodiment

[0719] The terminal provides a user interface for users to input queries. This user interface is implemented, for example, as a web browser or mobile application, allowing users to easily enter questions.

[0720] The terminal sends queries entered by the user to the server. These queries are sent using HTTP requests or other communication protocols. The terminal is also responsible for receiving the response from the server and displaying it to the user.

[0721] User-side embodiment

[0722] The user first enters a question about a specific law through the terminal's user interface. This query might be something like, "Please tell me how to convene a shareholders' meeting." The entered query is then sent to the server via the terminal.

[0723] The user checks the response sent from the server via their terminal. For example, a logical response such as, "In accordance with Article AA of the Companies Act, the following procedures must be followed to convene a shareholders' meeting," is displayed. In this way, users can quickly obtain accurate information regarding laws and regulations.

[0724] Specific example

[0725] For example, consider a case where a user enters the following query: "Please tell me about the procedures for establishing a company."

[0726] 1. The user enters the query using the terminal's user interface.

[0727] 2. The query is sent from the terminal to the server.

[0728] 3. The server receives the query and performs analysis.

[0729] 4. Use a generative artificial intelligence model to search the knowledge database and generate the optimal answer.

[0730] 5. The server sends the generated response to the terminal.

[0731] 6. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[0732] The above describes the embodiments of the present invention. This allows companies to efficiently handle complex legal compliance requirements and reduce significant effort and costs.

[0733] The following describes the processing flow.

[0734] Server-side processing steps

[0735] Step 1:

[0736] The server collects legal data. Specifically, it gathers legal texts, case studies, and expert opinions from the internet and legal databases, and inputs them into the system.

[0737] Step 2:

[0738] The server formats the collected legal data. This includes text normalization (e.g., standardizing uppercase and lowercase letters, removing unnecessary spaces, etc.) and data structuring (tagging and standardizing formatting).

[0739] Step 3:

[0740] The server trains a generative AI model using formatted legal data. The AI ​​model uses multi-layer neural networks and other methods to extract semantic information from the legal data and learn from it.

[0741] Step 4:

[0742] The server uses a trained generative AI model to build a knowledge database. Knowledge extraction primarily involves semantic analysis of text to create a foundation for generating legally compliant responses.

[0743] Step 5:

[0744] The server receives queries from users. These queries are received as text sent via HTTP requests, etc.

[0745] Step 6:

[0746] The server parses the received query. Based on the query's content, it searches the knowledge database to identify relevant legal provisions and explanations.

[0747] Step 7:

[0748] The server generates logical answers from relevant legal information. Using an AI model, it provides the most appropriate information in a sound and logical manner to user queries.

[0749] Step 8:

[0750] The server sends the generated response to the terminal. This also uses protocols such as HTTP responses to return data to the terminal.

[0751] Terminal-side processing steps

[0752] Step 1:

[0753] The terminal accepts query input from the user. This process takes place through the user interface of a web browser or mobile app.

[0754] Step 2:

[0755] The terminal sends user-entered queries to the server. Specifically, it sends data to the server using HTTP requests and API calls.

[0756] Step 3:

[0757] The terminal receives a response from the server. It analyzes the response data sent from the server and displays it appropriately in the user interface.

[0758] Step 4:

[0759] The terminal displays the response received from the server to the user. Text formatting and GUI elements are used to ensure the response is displayed clearly and understandably.

[0760] User-side processing steps

[0761] Step 1:

[0762] Users enter queries related to laws and regulations through the terminal's user interface. For example, they might enter a specific question such as, "Please tell me how to convene a shareholders' meeting."

[0763] Step 2:

[0764] The user sends the entered query to the server via their terminal. This ensures that the user's question reaches the server.

[0765] Step 3:

[0766] The user checks the response from the server displayed on their terminal. They view explanations of laws and procedures displayed on their terminal and use them in their actual work as needed. For example, they check the procedures for company establishment and follow those steps.

[0767] (Example 1)

[0768] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0769] In corporate legal affairs, it is crucial to quickly and accurately understand and implement complex regulatory provisions and procedures. However, legal documents themselves are vast in volume, making it difficult to extract necessary information and make appropriate judgments. Furthermore, considering expert opinions and past precedents is time-consuming and costly. Moreover, the scope of law is broad, making it difficult for the average corporate employee to deeply understand the regulations. There is a need for methods to solve these challenges and streamline corporate legal affairs.

[0770] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0771] In this invention, the server includes means for learning regulatory data using a generative artificial intelligence model and building a knowledge base; means for receiving regulatory questions from users; means for referencing appropriate regulatory statements from the knowledge base based on the questions and generating logical answers; means for providing the generated answers to users; means for collecting regulatory data and formatting it using natural language processing technology; means for generating answers from regulatory data using a generative AI model; and means for transmitting the generated answers to a terminal using a communication protocol. This makes it possible to quickly and accurately extract complex information regarding laws and regulations and provide users with appropriate answers.

[0772] A "generative artificial intelligence model" is an algorithm that learns from existing data and generates new information or text based on that information.

[0773] "Regulatory data" refers to legal information, including laws, regulations, precedents, and expert opinions.

[0774] A "knowledge base" is a database that systematically organizes and stores knowledge related to a specific field.

[0775] A "user" is someone who uses the system to input questions and receive answers.

[0776] A "question" is a specific information request that a user makes to a knowledge base.

[0777] A "regulatory document" is an official document or article based on laws and regulations.

[0778] A "logical answer" is a response that includes appropriate information and legal explanations in response to a question, and is logically consistent.

[0779] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and generate human language.

[0780] A "communication protocol" is a set of prescribed procedures or methods for sending and receiving data over a network.

[0781] A "terminal" is a device used by users to access a system, input questions, and receive answers.

[0782] This invention relates to a system that efficiently processes regulatory data using a generative AI model to support companies' legal compliance. This system operates through the cooperation of the server, terminal, and user.

[0783] Server-side embodiment

[0784] The server first collects and formats regulatory data. Specifically, it automatically collects data such as legal documents, case precedents, and expert opinions using web scraping and APIs. This task utilizes Python libraries such as BeautifulSoup and Scrapy.

[0785] The collected data is formatted using natural language processing (NLP) techniques. Python libraries such as spaCy and NLTK are used for text analysis and information extraction. As a result, the data is structured and stored in a knowledge base.

[0786] Next, the server trains the data using a generative AI model. Specifically, generative AI models such as GPT-4 are used. The dataset is input into the model, and it is trained using a multi-layer neural network. Data preprocessing and tokenization are also performed during this process.

[0787] After training, the server receives queries from users and performs analysis. Queries are sent to the server as HTTP requests, and natural language processing techniques are used for analysis. Based on the analysis results, the server searches a knowledge base and generates the most appropriate legal text or answer.

[0788] Terminal-side embodiment

[0789] The terminal provides an interface for the user to enter queries. This interface is implemented as a web browser or mobile application. The user uses the terminal to enter specific questions, which are then sent to the server as HTTP requests.

[0790] The terminal receives responses from the server and displays them in the user interface. This allows the user to obtain information quickly and accurately.

[0791] User-side embodiment

[0792] The user first enters a specific question through the terminal's user interface. For example, a question such as, "Please tell me how to convene a shareholders' meeting." The entered query is sent to the server via the terminal.

[0793] The user checks the response sent from the server on their device. For example, a logical response such as, "In accordance with Article AA of the Companies Act, the following procedures must be followed to convene a shareholders' meeting," will be displayed.

[0794] Specific example

[0795] For example, consider a case where a user enters the query, "Please tell me about the procedures for establishing a company."

[0796] 1. The user enters the query using the terminal's user interface.

[0797] 2. The query is sent from the terminal to the server.

[0798] 3. The server receives the query and performs analysis.

[0799] 4. Use a generative artificial intelligence model (e.g., GPT-4) to search the knowledge base and generate the optimal answer.

[0800] 5. The server sends the generated response to the terminal.

[0801] 6. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[0802] Thus, the present invention is a system that provides complex regulatory information quickly and accurately, thereby streamlining corporate legal compliance.

[0803] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0804] Step 1: Collection and formatting of regulatory data

[0805] The server automatically collects data such as legal documents, case precedents, and expert opinions using web scraping and APIs. Specifically, it uses Python libraries such as BeautifulSoup and Scrapy. The collected data is analyzed using natural language processing techniques with Python libraries such as spaCy and NLTK to extract important information. This generates structured data, which is then stored in a knowledge base.

[0806] Input: Legal documents, case law, expert opinions

[0807] Output: Structured regulatory data

[0808] Step 2: Training the Generative AI Model

[0809] The server trains a generative AI model (e.g., GPT-4) using regulatory data stored in its knowledge base. Specifically, it inputs the dataset into a multi-layer neural network model and trains it through text generation tasks. This process also includes data preprocessing and tokenization. This enables the model to generate appropriate answers to questions.

[0810] Input: Structured regulatory data

[0811] Output: Trained generative AI model

[0812] Step 3: Receive and parse queries

[0813] The terminal sends user queries to the server in the form of HTTP requests. The server analyzes the received queries using natural language processing techniques to extract topics and related keywords. This identifies the intent of the query.

[0814] Input: Query from user

[0815] Output: Analyzed query information (topics and keywords)

[0816] Step 4: Knowledge base search and answer generation

[0817] Based on the analyzed query information, the server searches its knowledge base to identify relevant legal texts and precedents. Next, the server uses a generative AI model to generate a logical response. Specifically, it creates an answer tailored to the user's query based on the identified legal texts and precedents.

[0818] Input: Parsed query information

[0819] Output: Generated logical answer

[0820] Step 5: Submit and view your response

[0821] The server sends the generated response to the terminal as an HTTP response. The terminal displays the received response in the user interface. This allows the user to immediately confirm the appropriate answer to the question.

[0822] Input: Generated logical answer

[0823] Output: Answer displayed in the user interface

[0824] (Application Example 1)

[0825] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0826] In industries such as electronic payment services, legal compliance is extremely important. However, laws are complex, and keeping up-to-date requires considerable effort and time. This makes it difficult for legal departments and employees to respond quickly and accurately to legal-related questions they face on a daily basis. To solve this problem, there is a need for a system that efficiently learns legal data and provides appropriate answers to user questions immediately.

[0827] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0828] This invention includes a server comprising means for learning legal data using a generative artificial intelligence model and constructing a knowledge database, means for receiving legal queries from users, means for referencing appropriate legal provisions from the knowledge database based on the queries and generating logical answers, means for providing the generated answers to users, and means for immediately answering legal-related questions in electronic payment services using an application installed on a smartphone. This enables electronic payment service providers to comply with laws and regulations quickly and accurately and reduce legal risks.

[0829] A "generative artificial intelligence model" is a type of artificial intelligence that has the ability to learn from large amounts of data and generate new information or answers based on that data.

[0830] "Legal data" refers to a dataset containing legal information and norms, including legal texts, case studies, and expert opinions.

[0831] A "knowledge database" is a database that organizes and stores information related to a specific area of ​​knowledge, making it quickly accessible.

[0832] A "query" is a question or request that a user enters in order to obtain specific information or an answer.

[0833] A "logical answer" is an accurate answer that is logically consistent and based on evidence.

[0834] "Means of providing information to users" refers to all methods and devices used to transmit information generated by the system to end users.

[0835] A "smartphone" is a small, portable information terminal that has information processing and communication capabilities.

[0836] "Electronic payment services" refer to all services that allow users to make payments using the internet or mobile devices.

[0837]

[0838] In order to implement this invention, the following system configuration and processing are necessary.

[0839] Server-side embodiment

[0840] The server first collects and formats legal data. This legal data includes legal text, case examples, and expert opinions. This data is used to train a generative AI model and build a knowledge database. Next, the server receives legal queries from users. These queries are sent in the form of specific questions from the user. Based on the query, the server searches the knowledge database, refers to the appropriate legal provisions, and generates a logical answer. The generated answer is provided to the user through an application installed on their smartphone.

[0841] Terminal-side embodiment

[0842] The terminal receives queries from the user through a user interface. This user interface is implemented, for example, as a web browser or mobile application, allowing users to easily input questions. The terminal sends the entered queries to the server. Queries are sent using communication protocols such as HTTP requests. The terminal also receives the response from the server and displays it to the user.

[0843] User-side embodiment

[0844] The user first enters a question about specific laws and regulations through the terminal's user interface. This query might be something like, "What are the legal requirements for KYC procedures?" The entered query is sent to the server via the terminal. The user then checks the answer sent from the server via the terminal, resulting in the rapid provision of accurate information about the laws and regulations.

[0845] Data processing and calculation

[0846] The following processes are performed on the server side:

[0847] Hardware: Servers and smartphones

[0848] Software: Python, Flask, HTTP communication library

[0849] First, a knowledge database is constructed by training an AI model with collected legal data. This knowledge database is designed to store the latest and most appropriate information regarding laws and regulations. Next, semantic information is extracted from the knowledge database to generate answers to user queries. Logical answers are generated through data retrieval and semantic information extraction based on the queries.

[0850] Specific example

[0851] For example, if a user enters the query "What are the legal requirements for KYC procedures for electronic payment services?" into an application when adding a new feature to an electronic payment platform for a fintech service, that query will be sent directly to the AI ​​model with a prompt like the following.

[0852] Example of a prompt:

[0853] "Could you please explain the legal requirements in Japan regarding KYC procedures?"

[0854] The above system enables electronic payment service providers to comply with laws and regulations quickly and accurately, thereby reducing legal risks.

[0855] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0856] Step 1:

[0857] The user enters specific questions (queries) regarding laws and regulations through the terminal's user interface. These queries might take the form of, for example, "What are the legal requirements for KYC procedures for electronic payment services?" The entered queries are collected by the terminal.

[0858] Step 2:

[0859] The terminal sends the collected queries to the server using communication protocols such as HTTP requests. During this process, the queries are formatted appropriately and sent in a way that makes them easy for the server to parse. The input is the user's query, and the output is the request containing that query.

[0860] Step 3:

[0861] The server parses the received queries. Natural language processing techniques are used for the analysis to extract the intent and keywords of the queries. Specifically, it identifies important information such as KYC procedures and legal requirements from the queries. The input is the query from the terminal, and the output is structured data with the analysis results.

[0862] Step 4:

[0863] The server searches the knowledge database based on the analysis results. The knowledge database contains legal data, and a generative AI model is used to select the appropriate legal provisions and explanations. The input is the analysis results, and the output is the optimal legal information. Keyword matching and prediction results from machine learning models are used for the search.

[0864] Step 5:

[0865] Using a generative AI model, logical answers are generated based on information extracted from a knowledge database. These answers include direct and specific explanations for the user's query. The input is legal information from the knowledge database, and the output is the generated answer. Text generation algorithms and transformer models are used in the generation process.

[0866] Step 6:

[0867] The server then sends the generated response back to the terminal. An HTTP response is used for transmission, ensuring the information is sent to the terminal in the appropriate format. The input is the generated response, and the output is the response containing the response.

[0868] Step 7:

[0869] The terminal displays the received response to the user through a user interface. Based on this response, the user can quickly obtain specific information regarding the law. The input is the response from the server, and the output is the information displayed on the user's screen.

[0870] Through the steps described above, users can quickly and accurately obtain specific information for compliance with laws and regulations regarding electronic payment services.

[0871] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0872] This invention combines a system that uses a generative artificial intelligence model to learn legal data and build a knowledge database with an emotion engine that recognizes user emotions, thereby enabling the provision of legal information tailored to the user's emotions. The following describes how the server, terminal, and user interact to realize the system.

[0873] Server-side embodiment

[0874] The server first collects and formats legal data. This legal data includes legal text, case studies, and expert opinions, which are then formatted appropriately. The formatted data is then aggregated into a knowledge database using a generative AI model.

[0875] Next, the server trains a generative artificial intelligence model. This training is performed using collected legal data, and it is a process in which the AI ​​model extracts and learns semantic information from the legal data. The trained model has the ability to generate appropriate legal information from the knowledge database.

[0876] The server also receives and analyzes queries from users. These queries relate to company law and are submitted by users in the form of specific questions. Based on the queries, the server searches its knowledge database, identifies relevant legal provisions and explanations, and generates answers. Furthermore, it uses an emotion engine to analyze the user's emotions and adjust the tone and expression of the answers accordingly.

[0877] Terminal-side embodiment

[0878] The terminal provides a user interface for receiving query input from the user. This user interface is implemented, for example, as a web browser or mobile application, allowing the user to easily input questions.

[0879] The terminal sends queries entered by the user to the server. Queries are sent using HTTP requests or other communication protocols. The terminal also receives responses from the server and displays them to the user. The displayed responses are generated by the server and adjusted by the sentiment engine.

[0880] User-side embodiment

[0881] The user first enters a question about a specific law through the terminal's user interface. This query might be something like, "Please tell me how to convene a shareholders' meeting." The entered query is then sent to the server via the terminal.

[0882] The server's response is analyzed by an emotion engine and then displayed on the terminal with adjusted tone and expression to match the user's emotions. The user then reviews the displayed legal explanations and procedures and uses them in their actual work.

[0883] Specific example

[0884] Consider a scenario where a user enters the following query: "Please tell me about the procedures for establishing a company."

[0885] 1. The user enters the query using the terminal's user interface.

[0886] 2. The query is sent from the terminal to the server.

[0887] 3. The server receives the query and performs analysis.

[0888] 4. Use a generative artificial intelligence model to search the knowledge database and generate the optimal answer.

[0889] 5. The emotion engine analyzes the user's emotions at the time of input and adjusts the tone and expression of the response as needed. For example, if it determines that the user is nervous, it will generate a response in a calmer and more friendly tone.

[0890] 6. The server sends the generated response to the terminal.

[0891] 7. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[0892] This invention enables companies to respond quickly and appropriately to complex legal issues, as well as to provide more user-friendly and easily understandable information that resonates with users' emotions.

[0893] The following describes the processing flow.

[0894] Server-side processing steps

[0895] Step 1:

[0896] The server collects legal data. Specifically, it obtains legal texts, case studies, and expert opinions from the internet and legal databases, and inputs them into the system.

[0897] Step 2:

[0898] The server formats the collected legal data. This includes text normalization (unifying uppercase and lowercase letters, removing unnecessary spaces, etc.) and data structuring (converting to a specific format).

[0899] Step 3:

[0900] The server trains a generative AI model using formatted legal data. The AI ​​model extracts semantic information from the legal data and learns how to generate text.

[0901] Step 4:

[0902] The server uses a trained generative AI model to build a knowledge database. This knowledge database neatly stores various information related to company law.

[0903] Step 5:

[0904] The server receives queries from users. Queries are sent via HTTP requests or similar methods and parsed by the server.

[0905] Step 6:

[0906] The server searches the knowledge database based on the received query, identifying relevant legal provisions and case examples.

[0907] Step 7:

[0908] The server uses an emotion engine to analyze the user's emotions. It recognizes the user's emotions (e.g., tension, excitement, questioning, etc.) from the context and expression of the query.

[0909] Step 8:

[0910] The server adjusts the tone and expression of its response based on the emotion recognition results to generate the most appropriate response. For example, if the user is nervous, it will use a calm and encouraging tone.

[0911] Step 9:

[0912] The server sends the generated response to the terminal. It returns data to the terminal using protocols such as HTTP responses.

[0913] Terminal-side processing steps

[0914] Step 1:

[0915] The terminal accepts query input from the user. Input is received through a user interface (e.g., a web browser or mobile app).

[0916] Step 2:

[0917] The terminal sends the query entered by the user to the server. Data is transmitted via HTTP requests and API calls.

[0918] Step 3:

[0919] The terminal receives a response from the server. It analyzes the response data sent from the server and prepares it for display on the user interface.

[0920] Step 4:

[0921] The terminal displays the user the response received from the server. The response is presented in an easy-to-read format using text formatting and GUI elements.

[0922] User-side processing steps

[0923] Step 1:

[0924] The user enters a query regarding the law through the terminal's user interface. For example, they might enter the question, "Please tell me how to convene a shareholders' meeting."

[0925] Step 2:

[0926] The user operates a terminal to send a query to the server. This sends the user's question to the system.

[0927] Step 3:

[0928] The user checks the response from the server displayed on their terminal. They view the legal explanations and procedural guides displayed on their terminal and use them in their actual work as needed. For example, they might check the specific procedures and required documents for company establishment.

[0929] Through the above processing flow, the system provides quick and appropriate answers to user questions and enables flexible responses that respond to the user's emotions.

[0930] (Example 2)

[0931] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0932] Conventional legal search systems can provide appropriate legal information when users ask questions about laws, but the information provided is often rigid and not user-friendly. Furthermore, they fail to provide information that takes into account the user's feelings and circumstances, which can cause users to feel anxious or stressed. This invention aims to provide a method for providing legal information that also takes into account the user's feelings.

[0933] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0934] In this invention, the server includes means for learning legal data using a generative artificial intelligence model and constructing a knowledge database; means for receiving legal queries from users; means for referencing appropriate legal provisions and related information from the knowledge database based on the queries and generating logical answers; means for sentiment analysis for adjusting the generated answers according to the user's emotions; and means for providing the adjusted answers to the user. This makes it possible to provide legal information in a format that takes the user's emotions into consideration.

[0935] A "generative artificial intelligence model" is an artificial intelligence system that extracts semantic information from various forms of data based on given training data and performs appropriate responses and inferences.

[0936] "Legal data" refers to information on norms and legal guidelines, including legal texts, case studies, and expert opinions.

[0937] A "knowledge database" is a database that aggregates legal data learned by a generative AI model and stores it in a searchable format.

[0938] "Emotion analysis methods" are technologies that analyze user input and behavior to identify their emotional state.

[0939] An "emotion engine" is a part of a system that analyzes the user's emotions and adjusts the output accordingly.

[0940] A "query" is a question or request that a user enters into a system.

[0941] A "user interface" is the interface through which a user interacts with a system, and includes web browsers and mobile applications.

[0942] A "logical response" is a response aimed at providing consistent and logically correct information in response to a user's query.

[0943] This invention is a system that provides users with appropriate legal information by utilizing a generative artificial intelligence model based on data related to laws and regulations. This system primarily operates through the cooperation of three entities: a server, a terminal, and a user.

[0944] Server-side embodiment

[0945] The server first collects and formats legal data. This data is gathered from the internet and dedicated databases. The collected data includes legal text, case examples, and expert opinions. The collected data is then formatted by normalizing the text and removing unnecessary tags and formatting.

[0946] Next, the server trains a generative AI model using the formatted data. Training is performed using libraries such as Python's TensorFlow and PyTorch. During the training process, the AI ​​model extracts semantic information from legal data and acquires the ability to interpret laws and infer related information. The trained model is then stored in a knowledge database.

[0947] When a user submits a query, the server parses it. Natural language processing (NLP) techniques are used for query parsing. Libraries such as spaCy and NLTK are used in this process. Based on the parsed query, the server searches a knowledge database. It identifies relevant legal provisions and explanations from the search results and generates an answer using a generative AI model. Furthermore, it analyzes the user's emotions using an emotion engine and adjusts the tone and expression of the answer.

[0948] Terminal-side embodiment

[0949] The terminal provides a user interface. This user interface is implemented as a web browser or mobile application, allowing users to easily input questions. The terminal sends the user-entered query to the server. An HTTP POST request is used for transmission. Once the server returns a response, it is displayed in the user interface. The displayed response is then refined by an emotion engine.

[0950] User-side embodiment

[0951] The user first enters a specific legal question through the terminal's user interface. For example, they might enter a question like, "Please tell me how to convene a shareholders' meeting." The entered query is sent to the server via the terminal. The server's response is analyzed by an emotion engine and displayed on the terminal with a tone and expression adjusted to match the user's emotions. The user then reviews the displayed explanation of the law and the procedures and uses it in their actual work.

[0952] Specific example

[0953] The specific processing flow when a user enters the query "Please tell me about the company incorporation procedure" is as follows:

[0954] 1. The user uses the terminal's user interface to enter the query, "Please tell me about the company incorporation procedure."

[0955] 2. The terminal sends a query to the server (HTTP POST request).

[0956] 3. The server receives the query and analyzes it using natural language processing. It then generates an answer from the knowledge database using a generative AI model.

[0957] 4. The emotion engine analyzes the user's emotions at the time of input and adjusts the tone and expression of the response. For example, if tension is detected, it will display, "Please rest assured. The company incorporation procedure is as follows..."

[0958] 5. The server sends the generated response to the terminal.

[0959] 6. The device displays the answer in the user interface.

[0960] 7. The user reviews the displayed answer and uses it to say, "The necessary procedures for company establishment are as follows:..."

[0961] This program enables companies to respond quickly and appropriately to complex legal issues, and also allows them to provide information that is sensitive to the user's feelings.

[0962] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0963] Step 1: Collection and formatting of legal data

[0964] The server collects legal texts, case studies, and expert opinions from the internet and dedicated databases. The collected data undergoes formatting processes such as text normalization and removal of unnecessary tags and formatting. Examples include the removal of HTML tags and classification by specific keywords.

[0965] Input: Legal text, case examples, expert opinions

[0966] Output: Normalized and formatted legal data

[0967] Step 2: Training the generative artificial intelligence model

[0968] The server uses formatted legal data to train a generative AI model. TensorFlow and PyTorch are used for this training. The model learns patterns and semantic information from the data, developing the ability to interpret laws and infer related information.

[0969] Input: Formatted legal data

[0970] Output: Trained Generative AI Model

[0971] Step 3: Receiving User Queries

[0972] The terminal accepts legal queries from users through its user interface. Users might enter questions such as, "Please tell me how to convene a shareholders' meeting."

[0973] Input: User query

[0974] Output: Unprocessed queries sent to the server

[0975] Step 4: Analyze the query

[0976] The server receives queries sent from terminals and analyzes them using natural language processing (NLP) techniques. Libraries such as spaCy and NLTK are used for this analysis. The server understands the content of the query and converts it into appropriate search keys.

[0977] Input: Unprocessed query

[0978] Output: Parsed query

[0979] Step 5: Search the knowledge database

[0980] The server searches the knowledge database based on the analyzed query. Using a generative AI model, it identifies relevant legal provisions and explanations and generates the optimal answer.

[0981] Input: Parsed query

[0982] Output: Initial draft answer

[0983] Step 6: Emotional Analysis and Tone Adjustment

[0984] The server uses an emotion engine to analyze the user's emotions from their queries. If the user is anxious, it adjusts the tone of its responses to be calmer. For example, if the emotion analysis engine determines from the user's input that they are "anxious," it might add the phrase "Please rest assured."

[0985] Input: Initial draft answer

[0986] Output: Adjusted answer

[0987] Step 7: Providing responses to users

[0988] The server sends the adjusted response to the terminal. The terminal displays the response to the user through the user interface. This allows the user to check specific legal information and procedures.

[0989] Input: Adjusted answer

[0990] Output: Final answer displayed on the user interface

[0991] As described above, this system involves collaboration between the server, terminal, and user to provide legal information that takes user emotions into consideration.

[0992] (Application Example 2)

[0993] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0994] Traditionally, legal information systems have been able to generate logical answers to user queries, but they have not adequately considered user emotions when providing information. This may lead to decreased user satisfaction in emergencies or stressful situations. Furthermore, to alleviate user anxiety, the method of providing legal information needs to be improved. In particular, for urgent legal information related to safety measures, it is necessary to provide information in an appropriate tone that responds to user emotions.

[0995] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for learning legal data using a generative artificial intelligence model and constructing a knowledge database, means for providing a user interface equipped with an emotion engine that recognizes the user's emotions, means for receiving legal queries from the user and analyzing emotion data, means for referring to appropriate legal articles from the knowledge database based on the query and generating a logical response in a tone that matches the user's emotions, and means for providing the generated response to the user's terminal. This makes it possible to provide legal information that takes the user's emotions into consideration, enabling the provision of information that is familiar and reassuring to the user even in emergencies or stressful situations. Furthermore, this improves user satisfaction and achieves prompt and appropriate legal compliance.

[0996] A "generative artificial intelligence model" is an artificial intelligence algorithm that automatically learns from collected data and performs a specific task.

[0997] "Legal data" refers to data that includes information related to the law, such as laws, regulations, precedents, and expert opinions.

[0998] A "knowledge database" is a database that organizes knowledge extracted from legal data and formats it in a way that makes it easy to search and refer to.

[0999] An "emotion engine" is a software component that analyzes the user's emotional state and adjusts the tone and content of its responses based on that analysis.

[1000] A "user interface" is an interface that includes input and output means for a user to interact with a system.

[1001] A "query" is a question or request that a user enters into a system to obtain specific information.

[1002] "Emotional data" refers to data that indicates the emotional state of a user, extracted from their facial expressions, voice, and text.

[1003] A "logical answer" is an accurate and appropriate answer that is logically constructed based on legal data.

[1004] "Tone" refers to the way a generated response is expressed and spoken, and it is an element that is adjusted according to the user's emotions.

[1005] This invention relates to a system that uses a generative artificial intelligence model to learn from legal data and provides sentiment-sensitive legal information in response to user queries. This system consists of three components: a server, a terminal, and a user.

[1006] Server-side embodiment

[1007] The server first collects and formats legal data. This legal data includes legal texts, case precedents, and expert opinions. The formatted data is then aggregated into a knowledge database using a generative AI model.

[1008] Next, the server trains a generative artificial intelligence model (e.g., GPT-4). Training is performed based on collected legal data, and it is a process in which the AI ​​model extracts semantic information from the legal data and learns from it. The trained model has the ability to generate appropriate legal information from the knowledge database.

[1009] The server also receives and analyzes queries from users. These queries include sentiment data (e.g., facial expressions, voice data). Based on the queries, the server searches a knowledge database to identify relevant legal provisions and explanations, and uses a sentiment engine to adjust the tone of the generated responses.

[1010] Terminal-side embodiment

[1011] The device provides a user interface for receiving query input from the user. This user interface is implemented as a smartphone application, allowing users to input questions via voice or text. It also utilizes the device's built-in camera and microphone for emotion recognition.

[1012] Queries and sentiment data are sent to the server using HTTP requests and other communication protocols. Responses from the server are received in real time and displayed on the user interface in an adjusted format.

[1013] User-side embodiment

[1014] The user first enters a question through the terminal's user interface. For example, they might enter, "What legal action should I take if a suspicious person enters my home?" This query is then sent to the server via the terminal.

[1015] The response from the server is adjusted by an emotion engine to match the user's emotions in terms of tone and expression, and then resent to the device. Specifically, if the user is feeling anxious, the response will be provided in a calm tone, such as, "Don't worry. Please follow these steps to ensure your safety."

[1016] Example of a prompt

[1017] If the user is feeling tense and anxious, generate a gentle response regarding how to deal with a suspicious person. Keep it simple and specific, including examples of responses that are easy for the user to understand.

[1018] This system allows users to quickly obtain appropriate legal information tailored to their emotions, even in emergencies. This leads to improved user satisfaction and enables prompt and appropriate legal responses.

[1019] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1020] Step 1:

[1021] The user enters queries through the device's user interface. For example, they might enter a query via voice or text, such as, "What legal action should I take if a suspicious person enters my home?" The system acquires query text as input, along with voice and facial expression data for sentiment analysis. This data is collected using the device's camera and microphone.

[1022] Step 2:

[1023] The terminal receives the entered query and sentiment data and sends it to the server. The input data includes the user's voice and facial expression data, as well as the query text. This data is sent to the server via an HTTP request.

[1024] Step 3:

[1025] On the server side, the emotion recognition module first analyzes the audio data and facial expression data to recognize the user's emotions. Specifically, it uses OpenCV and TensorFlow to extract emotional features from the audio and analyze the facial expressions acquired from the camera. The output of the emotion recognition is the emotion the user is feeling (e.g., anxiety, tension).

[1026] Step 4:

[1027] The server uses a generative AI model to search the knowledge database based on the query text and refer to appropriate legal information. Specifically, the generative AI model (e.g., GPT-4) extracts relevant information from legal data and generates a logical response. The query text and knowledge database are required as input, and the output is the generated legal information response.

[1028] Step 5:

[1029] The emotion engine receives the response generated by the generative AI model. The emotion engine adjusts the tone and expression of the response based on the user's emotions. For example, if the user is feeling anxious, the tone of the response will be calmed and adjusted to something like, "Don't worry. Please follow the steps below." The input requires the response from the generative AI model and the emotion recognition result, and the output is the adjusted response.

[1030] Step 6:

[1031] The server sends the adjusted response to the device. The received response is formatted in a way that is easy for the user to understand and sent back to the device as an HTTP response.

[1032] Step 7:

[1033] The terminal displays the reconciled response received from the server on the user interface. This response may be provided not only as text, but also as voice response or in other formats. This allows users to receive familiar and reassuring information in real time.

[1034] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1035] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1036] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[1037] [Fourth Embodiment]

[1038] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1039] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1040] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1041] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[1042] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[1043] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[1044] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[1045] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[1046] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[1047] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1048] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1049] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[1050] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1051] This invention is a system that utilizes generative AI to streamline the complex provisions and procedures of company law, thereby supporting companies' legal compliance. The following describes how the server, terminal, and user entities cooperate to implement the system.

[1052] Server-side embodiment

[1053] The server first collects and formats legal data. This legal data includes legal text, case studies, and expert opinions, which are then formatted appropriately. The formatted data is then aggregated into a knowledge database using a generative instinct artificial intelligence model.

[1054] Next, the server trains a generative artificial intelligence model. This training is performed by training a generative AI model, such as a multi-layer neural network, using the collected legal data. After training, the model will have the ability to extract semantic information from the knowledge database and generate answers to queries.

[1055] The server also receives and analyzes queries from users. These queries relate to company law and are submitted by users in the form of specific questions. Based on the queries, the server searches its knowledge database and generates logical answers that include the most relevant legal provisions and explanations. It then sends the generated answers to the users.

[1056] Terminal-side embodiment

[1057] The terminal provides a user interface for users to input queries. This user interface is implemented, for example, as a web browser or mobile application, allowing users to easily enter questions.

[1058] The terminal sends queries entered by the user to the server. These queries are sent using HTTP requests or other communication protocols. The terminal is also responsible for receiving the response from the server and displaying it to the user.

[1059] User-side embodiment

[1060] The user first enters a question about a specific law through the terminal's user interface. This query might be something like, "Please tell me how to convene a shareholders' meeting." The entered query is then sent to the server via the terminal.

[1061] The user checks the response sent from the server via their terminal. For example, a logical response such as, "In accordance with Article AA of the Companies Act, the following procedures must be followed to convene a shareholders' meeting," is displayed. In this way, users can quickly obtain accurate information regarding laws and regulations.

[1062] Specific example

[1063] For example, consider a case where a user enters the following query: "Please tell me about the procedures for establishing a company."

[1064] 1. The user enters the query using the terminal's user interface.

[1065] 2. The query is sent from the terminal to the server.

[1066] 3. The server receives the query and performs analysis.

[1067] 4. Use a generative artificial intelligence model to search the knowledge database and generate the optimal answer.

[1068] 5. The server sends the generated response to the terminal.

[1069] 6. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[1070] The above describes the embodiments of the present invention. This allows companies to efficiently handle complex legal compliance requirements and reduce significant effort and costs.

[1071] The following describes the processing flow.

[1072] Server-side processing steps

[1073] Step 1:

[1074] The server collects legal data. Specifically, it gathers legal texts, case studies, and expert opinions from the internet and legal databases, and inputs them into the system.

[1075] Step 2:

[1076] The server formats the collected legal data. This includes text normalization (e.g., standardizing uppercase and lowercase letters, removing unnecessary spaces, etc.) and data structuring (tagging and standardizing formatting).

[1077] Step 3:

[1078] The server trains a generative AI model using formatted legal data. The AI ​​model uses multi-layer neural networks and other methods to extract semantic information from the legal data and learn from it.

[1079] Step 4:

[1080] The server uses a trained generative AI model to build a knowledge database. Knowledge extraction primarily involves semantic analysis of text to create a foundation for generating legally compliant responses.

[1081] Step 5:

[1082] The server receives queries from users. These queries are received as text sent via HTTP requests, etc.

[1083] Step 6:

[1084] The server parses the received query. Based on the query's content, it searches the knowledge database to identify relevant legal provisions and explanations.

[1085] Step 7:

[1086] The server generates logical answers from relevant legal information. Using an AI model, it provides the most appropriate information in a sound and logical manner to user queries.

[1087] Step 8:

[1088] The server sends the generated response to the terminal. This also uses protocols such as HTTP responses to return data to the terminal.

[1089] Terminal-side processing steps

[1090] Step 1:

[1091] The terminal accepts query input from the user. This process takes place through the user interface of a web browser or mobile app.

[1092] Step 2:

[1093] The terminal sends user-entered queries to the server. Specifically, it sends data to the server using HTTP requests and API calls.

[1094] Step 3:

[1095] The terminal receives a response from the server. It analyzes the response data sent from the server and displays it appropriately in the user interface.

[1096] Step 4:

[1097] The terminal displays the response received from the server to the user. Text formatting and GUI elements are used to ensure the response is displayed clearly and understandably.

[1098] User-side processing steps

[1099] Step 1:

[1100] Users enter queries related to laws and regulations through the terminal's user interface. For example, they might enter a specific question such as, "Please tell me how to convene a shareholders' meeting."

[1101] Step 2:

[1102] The user sends the entered query to the server via their terminal. This ensures that the user's question reaches the server.

[1103] Step 3:

[1104] The user checks the response from the server displayed on their terminal. They view explanations of laws and procedures displayed on their terminal and use them in their actual work as needed. For example, they check the procedures for company establishment and follow those steps.

[1105] (Example 1)

[1106] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1107] In corporate legal affairs, it is crucial to quickly and accurately understand and implement complex regulatory provisions and procedures. However, legal documents themselves are vast in volume, making it difficult to extract necessary information and make appropriate judgments. Furthermore, considering expert opinions and past precedents is time-consuming and costly. Moreover, the scope of law is broad, making it difficult for the average corporate employee to deeply understand the regulations. There is a need for methods to solve these challenges and streamline corporate legal affairs.

[1108] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[1109] In this invention, the server includes means for learning regulatory data using a generative artificial intelligence model and building a knowledge base; means for receiving regulatory questions from users; means for referencing appropriate regulatory statements from the knowledge base based on the questions and generating logical answers; means for providing the generated answers to users; means for collecting regulatory data and formatting it using natural language processing technology; means for generating answers from regulatory data using a generative AI model; and means for transmitting the generated answers to a terminal using a communication protocol. This makes it possible to quickly and accurately extract complex information regarding laws and regulations and provide users with appropriate answers.

[1110] A "generative artificial intelligence model" is an algorithm that learns from existing data and generates new information or text based on that information.

[1111] "Regulatory data" refers to legal information, including laws, regulations, precedents, and expert opinions.

[1112] A "knowledge base" is a database that systematically organizes and stores knowledge related to a specific field.

[1113] A "user" is someone who uses the system to input questions and receive answers.

[1114] A "question" is a specific information request that a user makes to a knowledge base.

[1115] A "regulatory document" is an official document or article based on laws and regulations.

[1116] A "logical answer" is a response that includes appropriate information and legal explanations in response to a question, and is logically consistent.

[1117] "Natural language processing technology" refers to the technology that enables computers to understand, analyze, and generate human language.

[1118] A "communication protocol" is a set of prescribed procedures or methods for sending and receiving data over a network.

[1119] A "terminal" is a device used by users to access a system, input questions, and receive answers.

[1120] This invention relates to a system that efficiently processes regulatory data using a generative AI model to support companies' legal compliance. This system operates through the cooperation of the server, terminal, and user.

[1121] Server-side embodiment

[1122] The server first collects and formats regulatory data. Specifically, it automatically collects data such as legal documents, case precedents, and expert opinions using web scraping and APIs. This task utilizes Python libraries such as BeautifulSoup and Scrapy.

[1123] The collected data is formatted using natural language processing (NLP) techniques. Python libraries such as spaCy and NLTK are used for text analysis and information extraction. As a result, the data is structured and stored in a knowledge base.

[1124] Next, the server trains the data using a generative AI model. Specifically, generative AI models such as GPT-4 are used. The dataset is input into the model, and it is trained using a multi-layer neural network. Data preprocessing and tokenization are also performed during this process.

[1125] After training, the server receives queries from users and performs analysis. Queries are sent to the server as HTTP requests, and natural language processing techniques are used for analysis. Based on the analysis results, the server searches a knowledge base and generates the most appropriate legal text or answer.

[1126] Terminal-side embodiment

[1127] The terminal provides an interface for the user to enter queries. This interface is implemented as a web browser or mobile application. The user uses the terminal to enter specific questions, which are then sent to the server as HTTP requests.

[1128] The terminal receives responses from the server and displays them in the user interface. This allows the user to obtain information quickly and accurately.

[1129] User-side embodiment

[1130] The user first enters a specific question through the terminal's user interface. For example, a question such as, "Please tell me how to convene a shareholders' meeting." The entered query is sent to the server via the terminal.

[1131] The user checks the response sent from the server on their device. For example, a logical response such as, "In accordance with Article AA of the Companies Act, the following procedures must be followed to convene a shareholders' meeting," will be displayed.

[1132] Specific example

[1133] For example, consider a case where a user enters the query, "Please tell me about the procedures for establishing a company."

[1134] 1. The user enters the query using the terminal's user interface.

[1135] 2. The query is sent from the terminal to the server.

[1136] 3. The server receives the query and performs analysis.

[1137] 4. Use a generative artificial intelligence model (e.g., GPT-4) to search the knowledge base and generate the optimal answer.

[1138] 5. The server sends the generated response to the terminal.

[1139] 6. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[1140] Thus, the present invention is a system that provides complex regulatory information quickly and accurately, thereby streamlining corporate legal compliance.

[1141] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1142] Step 1: Collection and formatting of regulatory data

[1143] The server automatically collects data such as legal documents, case precedents, and expert opinions using web scraping and APIs. Specifically, it uses Python libraries such as BeautifulSoup and Scrapy. The collected data is analyzed using natural language processing techniques with Python libraries such as spaCy and NLTK to extract important information. This generates structured data, which is then stored in a knowledge base.

[1144] Input: Legal documents, case law, expert opinions

[1145] Output: Structured regulatory data

[1146] Step 2: Training the Generative AI Model

[1147] The server trains a generative AI model (e.g., GPT-4) using regulatory data stored in its knowledge base. Specifically, it inputs the dataset into a multi-layer neural network model and trains it through text generation tasks. This process also includes data preprocessing and tokenization. This enables the model to generate appropriate answers to questions.

[1148] Input: Structured regulatory data

[1149] Output: Trained generative AI model

[1150] Step 3: Receive and parse queries

[1151] The terminal sends user queries to the server in the form of HTTP requests. The server analyzes the received queries using natural language processing techniques to extract topics and related keywords. This identifies the intent of the query.

[1152] Input: Query from user

[1153] Output: Analyzed query information (topics and keywords)

[1154] Step 4: Knowledge base search and answer generation

[1155] Based on the analyzed query information, the server searches its knowledge base to identify relevant legal texts and precedents. Next, the server uses a generative AI model to generate a logical response. Specifically, it creates an answer tailored to the user's query based on the identified legal texts and precedents.

[1156] Input: Parsed query information

[1157] Output: Generated logical answer

[1158] Step 5: Submit and view your response

[1159] The server sends the generated response to the terminal as an HTTP response. The terminal displays the received response in the user interface. This allows the user to immediately confirm the appropriate answer to the question.

[1160] Input: Generated logical answer

[1161] Output: Answer displayed in the user interface

[1162] (Application Example 1)

[1163] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1164] In industries such as electronic payment services, legal compliance is extremely important. However, laws are complex, and keeping up-to-date requires considerable effort and time. This makes it difficult for legal departments and employees to respond quickly and accurately to legal-related questions they face on a daily basis. To solve this problem, there is a need for a system that efficiently learns legal data and provides appropriate answers to user questions immediately.

[1165] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[1166] This invention includes a server comprising means for learning legal data using a generative artificial intelligence model and constructing a knowledge database, means for receiving legal queries from users, means for referencing appropriate legal provisions from the knowledge database based on the queries and generating logical answers, means for providing the generated answers to users, and means for immediately answering legal-related questions in electronic payment services using an application installed on a smartphone. This enables electronic payment service providers to comply with laws and regulations quickly and accurately and reduce legal risks.

[1167] A "generative artificial intelligence model" is a type of artificial intelligence that has the ability to learn from large amounts of data and generate new information or answers based on that data.

[1168] "Legal data" refers to a dataset containing legal information and norms, including legal texts, case studies, and expert opinions.

[1169] A "knowledge database" is a database that organizes and stores information related to a specific area of ​​knowledge, making it quickly accessible.

[1170] A "query" is a question or request that a user enters in order to obtain specific information or an answer.

[1171] A "logical answer" is an accurate answer that is logically consistent and based on evidence.

[1172] "Means of providing information to users" refers to all methods and devices used to transmit information generated by the system to end users.

[1173] A "smartphone" is a small, portable information terminal that has information processing and communication capabilities.

[1174] "Electronic payment services" refer to all services that allow users to make payments using the internet or mobile devices.

[1175]

[1176] In order to implement this invention, the following system configuration and processing are necessary.

[1177] Server-side embodiment

[1178] The server first collects and formats legal data. This legal data includes legal text, case examples, and expert opinions. This data is used to train a generative AI model and build a knowledge database. Next, the server receives legal queries from users. These queries are sent in the form of specific questions from the user. Based on the query, the server searches the knowledge database, refers to the appropriate legal provisions, and generates a logical answer. The generated answer is provided to the user through an application installed on their smartphone.

[1179] Terminal-side embodiment

[1180] The terminal receives queries from the user through a user interface. This user interface is implemented, for example, as a web browser or mobile application, allowing users to easily input questions. The terminal sends the entered queries to the server. Queries are sent using communication protocols such as HTTP requests. The terminal also receives the response from the server and displays it to the user.

[1181] User-side embodiment

[1182] The user first enters a question about specific laws and regulations through the terminal's user interface. This query might be something like, "What are the legal requirements for KYC procedures?" The entered query is sent to the server via the terminal. The user then checks the answer sent from the server via the terminal, resulting in the rapid provision of accurate information about the laws and regulations.

[1183] Data processing and calculation

[1184] The following processes are performed on the server side:

[1185] Hardware: Servers and smartphones

[1186] Software: Python, Flask, HTTP communication library

[1187] First, a knowledge database is constructed by training an AI model with collected legal data. This knowledge database is designed to store the latest and most appropriate information regarding laws and regulations. Next, semantic information is extracted from the knowledge database to generate answers to user queries. Logical answers are generated through data retrieval and semantic information extraction based on the queries.

[1188] Specific example

[1189] For example, if a user enters the query "What are the legal requirements for KYC procedures for electronic payment services?" into an application when adding a new feature to an electronic payment platform for a fintech service, that query will be sent directly to the AI ​​model with a prompt like the following.

[1190] Example of a prompt:

[1191] "Could you please explain the legal requirements in Japan regarding KYC procedures?"

[1192] The above system enables electronic payment service providers to comply with laws and regulations quickly and accurately, thereby reducing legal risks.

[1193] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1194] Step 1:

[1195] The user enters specific questions (queries) regarding laws and regulations through the terminal's user interface. These queries might take the form of, for example, "What are the legal requirements for KYC procedures for electronic payment services?" The entered queries are collected by the terminal.

[1196] Step 2:

[1197] The terminal sends the collected queries to the server using communication protocols such as HTTP requests. During this process, the queries are formatted appropriately and sent in a way that makes them easy for the server to parse. The input is the user's query, and the output is the request containing that query.

[1198] Step 3:

[1199] The server parses the received queries. Natural language processing techniques are used for the analysis to extract the intent and keywords of the queries. Specifically, it identifies important information such as KYC procedures and legal requirements from the queries. The input is the query from the terminal, and the output is structured data with the analysis results.

[1200] Step 4:

[1201] The server searches the knowledge database based on the analysis results. The knowledge database contains legal data, and a generative AI model is used to select the appropriate legal provisions and explanations. The input is the analysis results, and the output is the optimal legal information. Keyword matching and prediction results from machine learning models are used for the search.

[1202] Step 5:

[1203] Using a generative AI model, logical answers are generated based on information extracted from a knowledge database. These answers include direct and specific explanations for the user's query. The input is legal information from the knowledge database, and the output is the generated answer. Text generation algorithms and transformer models are used in the generation process.

[1204] Step 6:

[1205] The server then sends the generated response back to the terminal. An HTTP response is used for transmission, ensuring the information is sent to the terminal in the appropriate format. The input is the generated response, and the output is the response containing the response.

[1206] Step 7:

[1207] The terminal displays the received response to the user through a user interface. Based on this response, the user can quickly obtain specific information regarding the law. The input is the response from the server, and the output is the information displayed on the user's screen.

[1208] Through the steps described above, users can quickly and accurately obtain specific information for compliance with laws and regulations regarding electronic payment services.

[1209] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[1210] This invention combines a system that uses a generative artificial intelligence model to learn legal data and build a knowledge database with an emotion engine that recognizes user emotions, thereby enabling the provision of legal information tailored to the user's emotions. The following describes how the server, terminal, and user interact to realize the system.

[1211] Server-side embodiment

[1212] The server first collects and formats legal data. This legal data includes legal text, case studies, and expert opinions, which are then formatted appropriately. The formatted data is then aggregated into a knowledge database using a generative AI model.

[1213] Next, the server trains a generative artificial intelligence model. This training is performed using collected legal data, and it is a process in which the AI ​​model extracts and learns semantic information from the legal data. The trained model has the ability to generate appropriate legal information from the knowledge database.

[1214] The server also receives and analyzes queries from users. These queries relate to company law and are submitted by users in the form of specific questions. Based on the queries, the server searches its knowledge database, identifies relevant legal provisions and explanations, and generates answers. Furthermore, it uses an emotion engine to analyze the user's emotions and adjust the tone and expression of the answers accordingly.

[1215] Terminal-side embodiment

[1216] The terminal provides a user interface for receiving query input from the user. This user interface is implemented, for example, as a web browser or mobile application, allowing the user to easily input questions.

[1217] The terminal sends queries entered by the user to the server. Queries are sent using HTTP requests or other communication protocols. The terminal also receives responses from the server and displays them to the user. The displayed responses are generated by the server and adjusted by the sentiment engine.

[1218] User-side embodiment

[1219] The user first enters a question about a specific law through the terminal's user interface. This query might be something like, "Please tell me how to convene a shareholders' meeting." The entered query is then sent to the server via the terminal.

[1220] The server's response is analyzed by an emotion engine and then displayed on the terminal with adjusted tone and expression to match the user's emotions. The user then reviews the displayed legal explanations and procedures and uses them in their actual work.

[1221] Specific example

[1222] Consider a scenario where a user enters the following query: "Please tell me about the procedures for establishing a company."

[1223] 1. The user enters the query using the terminal's user interface.

[1224] 2. The query is sent from the terminal to the server.

[1225] 3. The server receives the query and performs analysis.

[1226] 4. Use a generative artificial intelligence model to search the knowledge database and generate the optimal answer.

[1227] 5. The emotion engine analyzes the user's emotions at the time of input and adjusts the tone and expression of the response as needed. For example, if it determines that the user is nervous, it will generate a response in a calmer and more friendly tone.

[1228] 6. The server sends the generated response to the terminal.

[1229] 7. The terminal displays the following response to the user: "The procedures required when establishing a company are as follows:..."

[1230] This invention enables companies to respond quickly and appropriately to complex legal issues, as well as to provide more user-friendly and easily understandable information that resonates with users' emotions.

[1231] The following describes the processing flow.

[1232] Server-side processing steps

[1233] Step 1:

[1234] The server collects legal data. Specifically, it obtains legal texts, case studies, and expert opinions from the internet and legal databases, and inputs them into the system.

[1235] Step 2:

[1236] The server formats the collected legal data. This includes text normalization (unifying uppercase and lowercase letters, removing unnecessary spaces, etc.) and data structuring (converting to a specific format).

[1237] Step 3:

[1238] The server trains a generative AI model using formatted legal data. The AI ​​model extracts semantic information from the legal data and learns how to generate text.

[1239] Step 4:

[1240] The server uses a trained generative AI model to build a knowledge database. This knowledge database neatly stores various information related to company law.

[1241] Step 5:

[1242] The server receives queries from users. Queries are sent via HTTP requests or similar methods and parsed by the server.

[1243] Step 6:

[1244] The server searches the knowledge database based on the received query, identifying relevant legal provisions and case examples.

[1245] Step 7:

[1246] The server uses an emotion engine to analyze the user's emotions. It recognizes the user's emotions (e.g., tension, excitement, questioning, etc.) from the context and expression of the query.

[1247] Step 8:

[1248] The server adjusts the tone and expression of its response based on the emotion recognition results to generate the most appropriate response. For example, if the user is nervous, it will use a calm and encouraging tone.

[1249] Step 9:

[1250] The server sends the generated response to the terminal. It returns data to the terminal using protocols such as HTTP responses.

[1251] Terminal-side processing steps

[1252] Step 1:

[1253] The terminal accepts query input from the user. Input is received through a user interface (e.g., a web browser or mobile app).

[1254] Step 2:

[1255] The terminal sends the query entered by the user to the server. Data is transmitted via HTTP requests and API calls.

[1256] Step 3:

[1257] The terminal receives a response from the server. It analyzes the response data sent from the server and prepares it for display on the user interface.

[1258] Step 4:

[1259] The terminal displays the user the response received from the server. The response is presented in an easy-to-read format using text formatting and GUI elements.

[1260] User-side processing steps

[1261] Step 1:

[1262] The user enters a query regarding the law through the terminal's user interface. For example, they might enter the question, "Please tell me how to convene a shareholders' meeting."

[1263] Step 2:

[1264] The user operates a terminal to send a query to the server. This sends the user's question to the system.

[1265] Step 3:

[1266] The user checks the response from the server displayed on their terminal. They view the legal explanations and procedural guides displayed on their terminal and use them in their actual work as needed. For example, they might check the specific procedures and required documents for company establishment.

[1267] Through the above processing flow, the system provides quick and appropriate answers to user questions and enables flexible responses that respond to the user's emotions.

[1268] (Example 2)

[1269] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1270] Conventional legal search systems can provide appropriate legal information when users ask questions about laws, but the information provided is often rigid and not user-friendly. Furthermore, they fail to provide information that takes into account the user's feelings and circumstances, which can cause users to feel anxious or stressed. This invention aims to provide a method for providing legal information that also takes into account the user's feelings.

[1271] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[1272] In this invention, the server includes means for learning legal data using a generative artificial intelligence model and constructing a knowledge database; means for receiving legal queries from users; means for referencing appropriate legal provisions and related information from the knowledge database based on the queries and generating logical answers; means for sentiment analysis for adjusting the generated answers according to the user's emotions; and means for providing the adjusted answers to the user. This makes it possible to provide legal information in a format that takes the user's emotions into consideration.

[1273] A "generative artificial intelligence model" is an artificial intelligence system that extracts semantic information from various forms of data based on given training data and performs appropriate responses and inferences.

[1274] "Legal data" refers to information on norms and legal guidelines, including legal texts, case studies, and expert opinions.

[1275] A "knowledge database" is a database that aggregates legal data learned by a generative AI model and stores it in a searchable format.

[1276] "Emotion analysis methods" are technologies that analyze user input and behavior to identify their emotional state.

[1277] An "emotion engine" is a part of a system that analyzes the user's emotions and adjusts the output accordingly.

[1278] A "query" is a question or request that a user enters into a system.

[1279] A "user interface" is the interface through which a user interacts with a system, and includes web browsers and mobile applications.

[1280] A "logical response" is a response aimed at providing consistent and logically correct information in response to a user's query.

[1281] This invention is a system that provides users with appropriate legal information by utilizing a generative artificial intelligence model based on data related to laws and regulations. This system primarily operates through the cooperation of three entities: a server, a terminal, and a user.

[1282] Server-side embodiment

[1283] The server first collects and formats legal data. This data is gathered from the internet and dedicated databases. The collected data includes legal text, case examples, and expert opinions. The collected data is then formatted by normalizing the text and removing unnecessary tags and formatting.

[1284] Next, the server trains a generative AI model using the formatted data. Training is performed using libraries such as Python's TensorFlow and PyTorch. During the training process, the AI ​​model extracts semantic information from legal data and acquires the ability to interpret laws and infer related information. The trained model is then stored in a knowledge database.

[1285] When a user submits a query, the server parses it. Natural language processing (NLP) techniques are used for query parsing. Libraries such as spaCy and NLTK are used in this process. Based on the parsed query, the server searches a knowledge database. It identifies relevant legal provisions and explanations from the search results and generates an answer using a generative AI model. Furthermore, it analyzes the user's emotions using an emotion engine and adjusts the tone and expression of the answer.

[1286] Terminal-side embodiment

[1287] The terminal provides a user interface. This user interface is implemented as a web browser or mobile application, allowing users to easily input questions. The terminal sends the user-entered query to the server. An HTTP POST request is used for transmission. Once the server returns a response, it is displayed in the user interface. The displayed response is then refined by an emotion engine.

[1288] User-side embodiment

[1289] The user first enters a specific legal question through the terminal's user interface. For example, they might enter a question like, "Please tell me how to convene a shareholders' meeting." The entered query is sent to the server via the terminal. The server's response is analyzed by an emotion engine and displayed on the terminal with a tone and expression adjusted to match the user's emotions. The user then reviews the displayed explanation of the law and the procedures and uses it in their actual work.

[1290] Specific example

[1291] The specific processing flow when a user enters the query "Please tell me about the company incorporation procedure" is as follows:

[1292] 1. The user uses the terminal's user interface to enter the query, "Please tell me about the company incorporation procedure."

[1293] 2. The terminal sends a query to the server (HTTP POST request).

[1294] 3. The server receives the query and analyzes it using natural language processing. It then generates an answer from the knowledge database using a generative AI model.

[1295] 4. The emotion engine analyzes the user's emotions at the time of input and adjusts the tone and expression of the response. For example, if tension is detected, it will display, "Please rest assured. The company incorporation procedure is as follows..."

[1296] 5. The server sends the generated response to the terminal.

[1297] 6. The device displays the answer in the user interface.

[1298] 7. The user reviews the displayed answer and uses it to say, "The necessary procedures for company establishment are as follows:..."

[1299] This program enables companies to respond quickly and appropriately to complex legal issues, and also allows them to provide information that is sensitive to the user's feelings.

[1300] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1301] Step 1: Collection and formatting of legal data

[1302] The server collects legal texts, case studies, and expert opinions from the internet and dedicated databases. The collected data undergoes formatting processes such as text normalization and removal of unnecessary tags and formatting. Examples include the removal of HTML tags and classification by specific keywords.

[1303] Input: Legal text, case examples, expert opinions

[1304] Output: Normalized and formatted legal data

[1305] Step 2: Training the generative artificial intelligence model

[1306] The server uses formatted legal data to train a generative AI model. TensorFlow and PyTorch are used for this training. The model learns patterns and semantic information from the data, developing the ability to interpret laws and infer related information.

[1307] Input: Formatted legal data

[1308] Output: Trained Generative AI Model

[1309] Step 3: Receiving User Queries

[1310] The terminal accepts legal queries from users through its user interface. Users might enter questions such as, "Please tell me how to convene a shareholders' meeting."

[1311] Input: User query

[1312] Output: Unprocessed queries sent to the server

[1313] Step 4: Analyze the query

[1314] The server receives queries sent from terminals and analyzes them using natural language processing (NLP) techniques. Libraries such as spaCy and NLTK are used for this analysis. The server understands the content of the query and converts it into appropriate search keys.

[1315] Input: Unprocessed query

[1316] Output: Parsed query

[1317] Step 5: Search the knowledge database

[1318] The server searches the knowledge database based on the analyzed query. Using a generative AI model, it identifies relevant legal provisions and explanations and generates the optimal answer.

[1319] Input: Parsed query

[1320] Output: Initial draft answer

[1321] Step 6: Emotional Analysis and Tone Adjustment

[1322] The server uses an emotion engine to analyze the user's emotions from their queries. If the user is anxious, it adjusts the tone of its responses to be calmer. For example, if the emotion analysis engine determines from the user's input that they are "anxious," it might add the phrase "Please rest assured."

[1323] Input: Initial draft answer

[1324] Output: Adjusted answer

[1325] Step 7: Providing responses to users

[1326] The server sends the adjusted response to the terminal. The terminal displays the response to the user through the user interface. This allows the user to check specific legal information and procedures.

[1327] Input: Adjusted answer

[1328] Output: Final answer displayed on the user interface

[1329] As described above, this system involves collaboration between the server, terminal, and user to provide legal information that takes user emotions into consideration.

[1330] (Application Example 2)

[1331] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[1332] Traditionally, legal information systems have been able to generate logical answers to user queries, but they have not adequately considered user emotions when providing information. This may lead to decreased user satisfaction in emergencies or stressful situations. Furthermore, to alleviate user anxiety, the method of providing legal information needs to be improved. In particular, for urgent legal information related to safety measures, it is necessary to provide information in an appropriate tone that responds to user emotions.

[1333] In Application Example 2, the specific processing performed by the specific processing unit 290 of the data processing device 12 is realized by the following means. In this invention, the server includes means for learning legal data using a generative artificial intelligence model and constructing a knowledge database, means for providing a user interface equipped with an emotion engine that recognizes the user's emotions, means for receiving legal queries from the user and analyzing emotion data, means for referring to appropriate legal articles from the knowledge database based on the query and generating a logical response in a tone that matches the user's emotions, and means for providing the generated response to the user's terminal. This makes it possible to provide legal information that takes the user's emotions into consideration, enabling the provision of information that is familiar and reassuring to the user even in emergencies or stressful situations. Furthermore, this improves user satisfaction and achieves prompt and appropriate legal compliance.

[1334] A "generative artificial intelligence model" is an artificial intelligence algorithm that automatically learns from collected data and performs a specific task.

[1335] "Legal data" refers to data that includes information related to the law, such as laws, regulations, precedents, and expert opinions.

[1336] A "knowledge database" is a database that organizes knowledge extracted from legal data and formats it in a way that makes it easy to search and refer to.

[1337] An "emotion engine" is a software component that analyzes the user's emotional state and adjusts the tone and content of its responses based on that analysis.

[1338] A "user interface" is an interface that includes input and output means for a user to interact with a system.

[1339] A "query" is a question or request that a user enters into a system to obtain specific information.

[1340] "Emotional data" refers to data that indicates the emotional state of a user, extracted from their facial expressions, voice, and text.

[1341] A "logical answer" is an accurate and appropriate answer that is logically constructed based on legal data.

[1342] "Tone" refers to the way a generated response is expressed and spoken, and it is an element that is adjusted according to the user's emotions.

[1343] This invention relates to a system that uses a generative artificial intelligence model to learn from legal data and provides sentiment-sensitive legal information in response to user queries. This system consists of three components: a server, a terminal, and a user.

[1344] Server-side embodiment

[1345] The server first collects and formats legal data. This legal data includes legal texts, case precedents, and expert opinions. The formatted data is then aggregated into a knowledge database using a generative AI model.

[1346] Next, the server trains a generative artificial intelligence model (e.g., GPT-4). Training is performed based on collected legal data, and it is a process in which the AI ​​model extracts semantic information from the legal data and learns from it. The trained model has the ability to generate appropriate legal information from the knowledge database.

[1347] The server also receives and analyzes queries from users. These queries include sentiment data (e.g., facial expressions, voice data). Based on the queries, the server searches a knowledge database to identify relevant legal provisions and explanations, and uses a sentiment engine to adjust the tone of the generated responses.

[1348] Terminal-side embodiment

[1349] The device provides a user interface for receiving query input from the user. This user interface is implemented as a smartphone application, allowing users to input questions via voice or text. It also utilizes the device's built-in camera and microphone for emotion recognition.

[1350] Queries and sentiment data are sent to the server using HTTP requests and other communication protocols. Responses from the server are received in real time and displayed on the user interface in an adjusted format.

[1351] User-side embodiment

[1352] The user first enters a question through the terminal's user interface. For example, they might enter, "What legal action should I take if a suspicious person enters my home?" This query is then sent to the server via the terminal.

[1353] The response from the server is adjusted by an emotion engine to match the user's emotions in terms of tone and expression, and then resent to the device. Specifically, if the user is feeling anxious, the response will be provided in a calm tone, such as, "Don't worry. Please follow these steps to ensure your safety."

[1354] Example of a prompt

[1355] If the user is feeling tense and anxious, generate a gentle response regarding how to deal with a suspicious person. Keep it simple and specific, including examples of responses that are easy for the user to understand.

[1356] This system allows users to quickly obtain appropriate legal information tailored to their emotions, even in emergencies. This leads to improved user satisfaction and enables prompt and appropriate legal responses.

[1357] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1358] Step 1:

[1359] The user enters queries through the device's user interface. For example, they might enter a query via voice or text, such as, "What legal action should I take if a suspicious person enters my home?" The system acquires query text as input, along with voice and facial expression data for sentiment analysis. This data is collected using the device's camera and microphone.

[1360] Step 2:

[1361] The terminal receives the entered query and sentiment data and sends it to the server. The input data includes the user's voice and facial expression data, as well as the query text. This data is sent to the server via an HTTP request.

[1362] Step 3:

[1363] On the server side, the emotion recognition module first analyzes the audio data and facial expression data to recognize the user's emotions. Specifically, it uses OpenCV and TensorFlow to extract emotional features from the audio and analyze the facial expressions acquired from the camera. The output of the emotion recognition is the emotion the user is feeling (e.g., anxiety, tension).

[1364] Step 4:

[1365] The server uses a generative AI model to search the knowledge database based on the query text and refer to appropriate legal information. Specifically, the generative AI model (e.g., GPT-4) extracts relevant information from legal data and generates a logical response. The query text and knowledge database are required as input, and the output is the generated legal information response.

[1366] Step 5:

[1367] The emotion engine receives the response generated by the generative AI model. The emotion engine adjusts the tone and expression of the response based on the user's emotions. For example, if the user is feeling anxious, the tone of the response will be calmed and adjusted to something like, "Don't worry. Please follow the steps below." The input requires the response from the generative AI model and the emotion recognition result, and the output is the adjusted response.

[1368] Step 6:

[1369] The server sends the adjusted response to the device. The received response is formatted in a way that is easy for the user to understand and sent back to the device as an HTTP response.

[1370] Step 7:

[1371] The terminal displays the reconciled response received from the server on the user interface. This response may be provided not only as text, but also as voice response or in other formats. This allows users to receive familiar and reassuring information in real time.

[1372] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[1373] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1374] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[1375] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1376] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[1377] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[1378] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[1379] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[1380] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[1381] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[1382] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[1383] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[1384] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[1385] 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.

[1386] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[1387] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[1388] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[1389] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[1390] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[1391] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[1392] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[1393] The following is further disclosed regarding the embodiments described above.

[1394] (Claim 1)

[1395] A method for building a knowledge database by learning legal data using a generative artificial intelligence model,

[1396] A means of receiving user queries regarding laws and regulations,

[1397] A means of referencing appropriate legal provisions from a knowledge database based on a query and generating a logical response,

[1398] A means of providing the generated answer to the user,

[1399] A system that includes this.

[1400] (Claim 2)

[1401] The system according to claim 1, further comprising means for collecting and formatting data including legal text, case examples, and expert opinions as legal data.

[1402] (Claim 3)

[1403] The system according to claim 1, further comprising means for analyzing user queries and providing generated answers in real time.

[1404] (Claim 4)

[1405] The system according to claim 1, wherein the means for learning legal data using a generative artificial intelligence model and constructing a knowledge database includes the processes of data formatting, model training, and knowledge extraction.

[1406] (Claim 5)

[1407] A means of allowing users to input queries using a user interface,

[1408] The system according to claim 1, further comprising means for sending queries from a user's terminal to a server.

[1409] "Example 1"

[1410] (Claim 1)

[1411] A means of building a knowledge base by learning regulatory data using a generative artificial intelligence model,

[1412] A means of receiving questions from users regarding regulations,

[1413] A means of referencing appropriate regulatory statements from a knowledge base based on a question and generating a logical answer,

[1414] A means of providing the generated response to the user,

[1415] A means of collecting regulatory data and formatting it using natural language processing technology,

[1416] A means of generating responses from regulatory data using a generative AI model,

[1417] A means of sending the generated response to the terminal using a communication protocol,

[1418] A system that includes this.

[1419] (Claim 2)

[1420] The system according to claim 1, further comprising means for collecting data including regulatory texts, case law, and expert opinions as regulatory data, and formatting it using natural language processing technology.

[1421] (Claim 3)

[1422] The system according to claim 1, further comprising means for analyzing questions from users and providing generated answers in real time.

[1423] "Application Example 1"

[1424] (Claim 1)

[1425] A method for building a knowledge database by learning legal data using a generative artificial intelligence model,

[1426] A means of receiving user queries regarding laws and regulations,

[1427] A means of referencing appropriate legal provisions from a knowledge database based on a query and generating a logical response,

[1428] A means of providing the generated answer to the user,

[1429] A means of instantly answering legal questions regarding electronic payment services using an application installed on a smartphone,

[1430] A system that includes this.

[1431] (Claim 2)

[1432] The system according to claim 1, further comprising means for collecting and formatting data including legal texts, case examples, and expert opinions as legal data.

[1433] (Claim 3)

[1434] The system according to claim 1, further comprising means for analyzing user queries and providing generated answers in real time.

[1435]

[1436] "Example 2 of combining an emotion engine"

[1437] (Claim 1)

[1438] A method for building a knowledge database by learning legal data using a generative artificial intelligence model,

[1439] A means of receiving user queries regarding laws and regulations,

[1440] A means of generating a logical response by referencing appropriate legal provisions and related information from a knowledge database based on a query,

[1441] A sentiment analysis method that adjusts the generated response according to the user's emotions,

[1442] A means of providing users with adjusted answers,

[1443] A system that includes this.

[1444] (Claim 2)

[1445] The system according to claim 1, further comprising means for collecting and formatting text data, case examples, and expert opinions as legal data.

[1446] (Claim 3)

[1447] The system according to claim 1, further comprising means for analyzing user queries and providing generated answers in real time.

[1448] "Application example 2 when combining with an emotional engine"

[1449] (Claim 1)

[1450] A method for building a knowledge database by learning legal data using a generative artificial intelligence model,

[1451] A means of providing a user interface equipped with an emotion engine that recognizes user emotions,

[1452] A means of receiving legal queries from users and analyzing sentiment data,

[1453] A means of referencing appropriate legal provisions from a knowledge database based on a query and generating a logical response in a tone that responds to the user's emotions,

[1454] A means of providing the generated response to the user's device,

[1455] A system that includes this.

[1456] (Claim 2)

[1457] The system according to claim 1, further comprising means for collecting and formatting data including legal text, case examples, and expert opinions as legal data.

[1458] (Claim 3)

[1459] The system according to claim 1, further comprising means for analyzing user queries and sentiment data and providing the generated responses in real time. [Explanation of Symbols]

[1460] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A method for building a knowledge database by learning legal data using a generative artificial intelligence model, A means of receiving user queries regarding laws and regulations, A means of referencing appropriate legal provisions from a knowledge database based on a query and generating a logical response, A means of providing the generated answer to the user, A system that includes this.

2. The system according to claim 1, further comprising means for collecting and formatting data including legal text, case examples, and expert opinions as legal data.

3. The system according to claim 1, further comprising means for analyzing user queries and providing generated answers in real time.

4. The system according to claim 1, wherein the means for learning legal data using a generative artificial intelligence model and constructing a knowledge database includes the processes of data formatting, model training, and knowledge extraction.

5. A means of allowing users to input queries using a user interface, The system according to claim 1, further comprising means for sending queries from a user's terminal to a server.

Images