System

Abstract

A system is provided.SOLUTION: The system includes a means for collecting case data of past civil and domestic lawsuits and storing the data in a database, a means for receiving lawsuit information inputted by a user and retrieving similar past cases from the database, and a means for calculating a winning rate of a lawsuit based on the retrieved similar cases and providing the result to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Currently, legal services are expensive, making it difficult for many people to receive the legal support they need. Litigation costs, particularly in civil and family cases, place a heavy burden on some families, potentially resulting in lost opportunities for fair justice. Furthermore, litigation requires a great deal of time and effort, and a lack of efficiency contributes to delays in legal proceedings. In addition, the inability to effectively utilize information on court records and lawyers' advocacy activities is another factor hindering improvements in the quality of litigation support. [Means for solving the problem]

[0005] This invention provides a system that includes a means for collecting and storing past civil and family law case data in a database, a means for receiving case information entered by a user and searching the database for similar past cases, and a means for calculating the success rate of a case based on the searched similar cases and providing the result to the user. The system also includes a means for collecting court records and lawyers' legal activity data and training a generating AI, a means for a user to input information for requesting legal assistance from the generating AI, and a means for the generating AI to conduct legal activities in court. This system reduces the cost and time required for litigation, creates an environment where ordinary households can easily receive legal assistance, and realizes efficient and high-quality litigation support. Additionally, by including a means for visually displaying the calculation results of the success rate and details of similar cases through a user interface and a means for providing advice to help users decide whether to proceed with litigation, the system provides users with clear and practical information to support their decision-making regarding litigation.

[0006] "Past civil and family law case data" refers to detailed information on past civil and family law case decisions and lawsuits that have been filed in accordance with the law.

[0007] A "database" is an information storage system that systematically organizes and stores collected case law data and related information, allowing for quick search and reference as needed.

[0008] "User" refers to an individual or organization that uses this system to input litigation information and obtain advice on winning rates and defense activities.

[0009] "Lawsuit Information" includes detailed information about the specifics of the lawsuit the user is about to file, such as the subject matter of the lawsuit, the desired outcome, and evidence.

[0010] "Receiving" refers to the act of the server receiving data or information sent from the user.

[0011] "Similar past cases" refers to past litigation cases that have a high degree of relevance or commonality when current litigation information is compared with past precedent data.

[0012] "Search" refers to the act of finding required data within a database based on specific conditions or criteria.

[0013] The "lawsuit success rate" is a statistical calculation of the likelihood of winning a current lawsuit based on the outcomes of past similar cases.

[0014] "Providing" refers to the act of submitting the calculated winning rate and related information to the user to help them understand.

[0015] "Court records" refers to all documents, minutes, defenses, evidence, etc. recorded during the actual trial.

[0016] "A lawyer's legal activities" refers to a series of legal procedures, arguments, presentation of evidence, and other activities that a lawyer carries out on behalf of a client in court.

[0017] "Generative AI" refers to artificial intelligence that learns from collected data and generates legal advice and recommendations.

[0018] "Courtroom advocacy" refers to all advocacy activities undertaken by the generated AI in court on behalf of a client.

[0019] "User interface" refers to the screen and operating means that allow a user to interact with a system. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0021] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0022] First, the terms used in the following description will be explained.

[0023] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0024] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0025] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0026] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0027] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0031] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0032] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0033] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0034] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0035] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0038] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0039] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0040] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0041] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data of civil and family law lawsuits, and furthermore, uses generative AI to conduct defense activities in court. This system operates in cooperation with three parties: a server, a terminal, and a user.

[0042] Data collection and storage

[0043] The server periodically collects case data for past civil and family law cases from legal-related public data sources and open databases. This data includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[0044] Entering lawsuit information and calculating win rate

[0045] The user uses the terminal to input information related to their legal case, including the specifics of the case, the desired outcome, and supporting documents. This input data is then sent from the terminal to the server.

[0046] The server searches the database based on the received user's lawsuit information to extract similar past cases. This is done using natural language processing (NLP) algorithms and similarity calculations. Based on the analysis results of the extracted similar cases, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the lawsuit. The calculation results are generated in report format and sent to the terminal.

[0047] The user can review the reports provided by the terminal and view details of the winning rate and related cases. Based on this information, the user can decide whether to proceed with the lawsuit.

[0048] Generative AI advocacy

[0049] When legal changes are made, the server collects data on court records and lawyers' legal activities and trains the generative AI. This training data includes minutes, evidence, and legal content. The generative AI acquires legal skills to support litigation based on the learned data.

[0050] When a user requests that the generated AI represent them, they input the necessary information into their device and send a request for the generated AI to the server. The server then uses this information and learning data to formulate an optimal defense strategy. When the generated AI represents its case in court, it provides real-time advice to the user and supports the progress of the trial.

[0051] Users can monitor the AI's defense activities in court and provide instructions as necessary. Once the outcome of the lawsuit is known, a detailed report will be provided via the device, along with advice on future actions.

[0052] Specific examples

[0053] For example, suppose a user is considering divorce proceedings and is fighting over asset division and child custody. The user uses a terminal to enter the details of the lawsuit and supporting documents. The server receives this information and searches its database for similar past divorce cases. Based on the search results, the server calculates a 70% chance of success and provides a report to the user.

[0054] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend their case. The generated AI will study past court records and conduct a defense in court. Finally, if the generated AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0055] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment in which ordinary households can easily receive legal support.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] The server collects historical civil and family law case data from legal public data sources and open databases, using web scraping and APIs to normalize and retrieve the data.

[0059] Step 2:

[0060] The server preprocesses the collected case data, converting it into a standard format before storing it in a database, including deduplication, filling in missing values, and text cleansing.

[0061] Step 3:

[0062] The user uses a terminal to enter information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents). The input form has required and optional fields, and the user must enter all required information.

[0063] Step 4:

[0064] The terminal transmits the entered litigation information to the server, where the transmitted data is encoded and sent using a secure communication method.

[0065] Step 5:

[0066] The server analyzes the received user's lawsuit information and searches for similar past cases in its database, using natural language processing (NLP) algorithms and similarity calculations.

[0067] Step 6:

[0068] The server calculates the probability of winning a lawsuit based on the data of similar cases retrieved, using machine learning models (such as logistic regression and random forests) to take into account necessary variables.

[0069] Step 7:

[0070] The server generates a report detailing the winning probability calculation and related similar cases, including a detailed explanation of the calculation process to ensure transparency.

[0071] Step 8:

[0072] The server sends the generated reports to the terminal, which are presented in a visually easy-to-understand format (graphs, tables, text).

[0073] Step 9:

[0074] The terminal displays the submitted report to the user through a user interface, where the user can view the win rate, similar cases, and specific advice.

[0075] Step 10:

[0076] Users can decide whether to proceed with a lawsuit based on the reports and advice provided, and can also ask additional questions or complete the data.

[0077] Step 11:

[0078] When legal changes are made, the server collects data on court records and lawyers' legal activities, including minutes, evidence, and content of legal arguments, and organizes it into a learning dataset.

[0079] Step 12:

[0080] The server trains the generative AI using data based on court records, using deep learning techniques to train the AI ​​to acquire effective advocacy skills.

[0081] Step 13:

[0082] To request the generated AI to defend the case, the user re-enters the necessary legal information into the device and submits the request.

[0083] Step 14:

[0084] The server receives a request for legal representation from the user and formulates an optimal legal representation strategy based on the learning data, and if necessary, communicates with the user.

[0085] Step 15:

[0086] Once litigation begins, the generative AI will actually act as a defense in court, providing real-time advice to the user and supporting the proceedings, including suggesting when to present evidence and what questions to ask.

[0087] Step 16:

[0088] After the verdict is reached, the server analyzes the results and generates a detailed report that is sent to the device, including the outcome of the verdict and advice on future actions.

[0089] Step 17:

[0090] The user will use the provided report to determine any necessary follow-up actions (eg, appeal decisions, etc.).

[0091] The above are the specific processing steps of the entire system.

[0092] Example 1

[0093] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0094] Existing legal support systems have difficulty calculating an appropriate win rate for each user's legal information, or providing real-time support for legal activities based on past cases. Furthermore, to efficiently support legal activities in court, it is necessary to analyze a huge amount of past court records and lawyers' legal activity data, and use generative AI to formulate optimal legal strategies.

[0095] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0096] In this invention, the server includes means for collecting past litigation case data from legal-related public data sources and open databases and storing it in a database, means for receiving litigation information entered by a user and searching the database for similar past cases using a natural language processing algorithm, means for analyzing the searched similar cases and calculating the success rate of the litigation using a machine learning model and providing the results to the user in the form of a report, and means for the user to use a terminal for inputting litigation information and evidential materials. This allows users to calculate the success rate of their own litigation cases with high accuracy and to develop optimal defense strategies based on similar past cases.

[0097] "Legal-related public data sources" refers to databases of laws, precedents, court records, etc. that are publicly available from governments and related agencies.

[0098] "Open database" refers to a database provided by a third party that is freely accessible on the Internet.

[0099] "Litigation case data" refers to a collection of data including records of past civil and family law litigation, judgments, related laws and regulations, evidence lists, and background information on the parties involved.

[0100] "Database" refers to an information system for efficiently storing, retrieving, and managing information.

[0101] "Law case information entered by the user" refers to the specific details, desired outcome, supporting documents, etc., of the user's own legal case.

[0102] "Natural language processing algorithm" refers to technology that allows computers to analyze human language and understand its meaning.

[0103] "Similar past cases" refers to past legal cases that have similar characteristics based on specific criteria to the user's legal information.

[0104] A "machine learning model" refers to a statistical algorithm that uses data to learn and make predictions, classifications, and other similar tasks.

[0105] "Generative AI" refers to artificial intelligence that can learn from past data and automatically perform specific tasks.

[0106] A "defense strategy" refers to a plan or method that aims to gain an advantage in a lawsuit based on law and evidence.

[0107] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data for civil and family law lawsuits, and uses generative AI to conduct defense activities in court. This system operates in cooperation with three parties: a server, a terminal, and a user.

[0108] Data collection and storage

[0109] The server periodically collects past civil and family law case data from legal-related public data sources and open databases using curl and APIs. The collected data is then stored in a database (e.g., MySQL (registered trademark) or PostgreSQL) after checking its accuracy. An index is also added to make the data easier to search and manage.

[0110] Entering lawsuit information and calculating win rate

[0111] The user enters information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents) into a dedicated form on the device. The device temporarily stores the entered data, and once all the data is collected, it sends an HTTP POST request to the server.

[0112] The server uses a natural language processing (NLP) algorithm to search its database for similar past cases based on the received user's lawsuit information. Specifically, it generates a search query and performs a full-text search of the past lawsuit database. Based on the analysis results of the extracted similar cases, it uses a machine learning model (e.g., logistic regression or random forest) to calculate the success rate of the lawsuit.

[0113] The results are formatted into a report and sent from the server to the terminal. The user can then review the report provided by the terminal and refer to the winning rate and details of related cases. Based on this information, the user can decide whether to proceed with the lawsuit.

[0114] Generative AI advocacy

[0115] When legal changes or new legal precedents arise, the server collects court records and lawyers' advocacy data and trains the generative AI using a natural language processing framework (e.g., BERT). This training data includes court records, evidence, and advocacy content.

[0116] When a user requests a defense from the generated AI, they input the necessary information into their device and send a defense request to the server. The server receives this information and provides the necessary data to the generated AI.

[0117] The generative AI will develop an optimal defense strategy based on past learning data and support legal actions in court. It will send advice to the user in real time to support the trial. The user can monitor the generative AI's legal actions in court and give instructions as necessary. Once the outcome of the trial is known, a detailed report will be provided to the user via their device, along with advice on future countermeasures.

[0118] Specific examples

[0119] For example, suppose a user is considering divorce proceedings and is fighting over asset division and child custody. The user uses a terminal to enter the details of the lawsuit and supporting documents. The server receives this information and searches its database for similar past divorce cases. Based on the search results, the server calculates a 70% chance of success and provides a report to the user.

[0120] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend their case. The generated AI will learn from past court records and conduct a defense in court. Finally, if the generated AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0121] An example prompt is, "I am currently in a divorce proceeding regarding asset division and child custody. How can I calculate my chances of winning based on past case data and use generative AI to defend my case in court?"

[0122] The above is a specific embodiment of the present invention. This system will reduce the cost and time required for litigation, and create an environment in which ordinary households can easily receive legal support.

[0123] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0124] Step 1:

[0125] The server periodically collects past civil and family law case data from legal-related public data sources and open databases via the Internet, using curl and APIs to retrieve the data.

[0126] Input: Data source URL or API endpoint

[0127] Data processing: Formatting the acquired data and checking its accuracy

[0128] Output: A formatted dataset

[0129] Specific example: "curl https: / / example-legal-database.com / api / cases"

[0130] Step 2:

[0131] The server stores the collected data in a database (e.g., MySQL or PostgreSQL) and indexes it for efficient searching.

[0132] Input: Formatted dataset

[0133] Data manipulation: Inserting and indexing data based on the database schema

[0134] Output: Data stored in the database

[0135] Example: "INSERT INTO cases_table (case_id, case_type, verdict, lAWS(registered trademark), evidence, parties_info) VALUES (...)"

[0136] Step 3:

[0137] The user enters the case information into a dedicated form on the terminal.

[0138] Input: Contents of lawsuit, desired outcome, evidence, etc.

[0139] Data calculation: Temporarily save input data

[0140] Output: Temporarily saved dataset

[0141] Specific example: "Enter the details of the lawsuit in text on a web form"

[0142] Step 4:

[0143] The terminal sends the completed input data to the server via an HTTP POST request.

[0144] Input: Completed dataset

[0145] Data processing: Generating the request body

[0146] Output: Request sent to server

[0147] Example: "POST / api / case_submission { case_data: {...}}"

[0148] Step 5:

[0149] Based on the user's lawsuit information received by the server, a natural language processing (NLP) algorithm is used to search for similar past cases.

[0150] Input: User's case information

[0151] Data Computing: Database Querying Using Full-Text Search Algorithms

[0152] Output: List of similar jobs

[0153] Specific example: SELECT FROM cases_table WHERE MATCH (case_text) AGAINST ('User's lawsuit details')

[0154] Step 6:

[0155] The server analyzes data from similar cases and uses machine learning models (such as logistic regression or random forest) to calculate the probability of winning a lawsuit.

[0156] Input: Data for similar cases

[0157] Data Computation: Probability Computation Using Machine Learning Models

[0158] Output: Win rate calculation result

[0159] Specific example: "Calculating probability using a machine learning model (e.g., sklearn's LogisticRegression.predict_proba)"

[0160] Step 7:

[0161] The server formats the winning rate calculation results into a report format and sends it to the terminal.

[0162] Input: Win rate calculation result

[0163] Data processing: Report generation (e.g. HTML or PDF format)

[0164] Output: Report sent to user terminal

[0165] Specific example: "Using a report generation library (e.g. ReportLab)"

[0166] Step 8:

[0167] The user reviews the report on the terminal and decides whether to proceed with the lawsuit.

[0168] Input: Report data

[0169] Data calculation: Understanding and evaluating report content

[0170] Output: Decision on next action

[0171] Specific example: "Display report PDF on browser"

[0172] Step 9:

[0173] The user inputs the necessary information to request legal representation from the generated AI and sends it from the terminal to the server.

[0174] Input: Detailed information about the legal request

[0175] Data processing: Generating request data

[0176] Output: Request sent to server

[0177] Specific example: "Enter legal request information on a web form"

[0178] Step 10:

[0179] The server provides the received request information to the generation AI, which then formulates the optimal defense strategy.

[0180] Input: Legal request information

[0181] Data Computation: Strategy Planning with Generative AI

[0182] Output: Strategic proposal

[0183] Specific example: "API call to pass request information to the generation AI"

[0184] Step 11:

[0185] Generative AI will act as a defense in court and provide real-time advice to users.

[0186] Input: Strategic data generated by generative AI

[0187] Data Computing: Real-time Advice Generation

[0188] Output: Provide advice to the user

[0189] Specific example of operation: "Generative AI sends advice in real time in chat format"

[0190] Step 12:

[0191] The user will monitor the generated AI's defense activities in court and provide instructions as necessary.

[0192] Input: Generative AI advice and on-site situation

[0193] Data Computing: Advice and Situation Assessment

[0194] Output: Sending instructions

[0195] Specific example: "Real-time feedback and instructions on web applications"

[0196] Step 13:

[0197] When the outcome of the lawsuit becomes clear, the server generates a detailed report and sends it to the device, along with advice on future countermeasures.

[0198] Input: Lawsuit outcome data

[0199] Data processing: detailed report generation and analysis of future countermeasures

[0200] Output: Detailed report and further advice to the user

[0201] Specific example: "Generating and sending a final result report"

[0202] (Application example 1)

[0203] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0204] In the past, in civil and family law cases, it was difficult to calculate the success rate of a case based on past case data, and there was a lack of means to provide real-time advice to properly support legal advocacy in court. As a result, in many cases, users found it difficult to make informed decisions, and the quality of legal advocacy was inconsistent.

[0205] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0206] In this invention, the server includes means for collecting case data on past civil and family law litigation cases and storing it in a database, means for receiving litigation information entered by a user and searching the database for similar past cases, means for calculating the success rate of the litigation based on the searched similar cases and providing the result to the user in real time, and means for receiving the litigation information entered by a user via a smartphone and calculating the success rate of the litigation in real time. This allows the user to obtain a highly accurate litigation success rate based on past cases and to obtain the necessary information in real time via their smartphone.

[0207] "Past civil and family law case precedent data" refers to past civil and family law case precedent information collected from public institutions and legal databases.

[0208] "Means of storing in a database" refers to a method or system for systematically organizing collected case law data and storing it so that it can be easily searched and analyzed.

[0209] "Means for receiving litigation information entered by the user and searching for similar past cases" refers to technologies and algorithms for obtaining information about litigation provided by the user and using that information to find similar cases from past case data.

[0210] "Means for calculating the success rate of a lawsuit and providing the result to the user in real time" refers to technology or methods that analyze past case law data and information provided by the user to calculate the likelihood that the outcome of the lawsuit will be favorable to the user and present the result to the user in real time.

[0211] "Means of obtaining necessary information in real time via a smartphone" refers to applications and systems that allow users to receive information provided in real time via a smartphone, which is a mobile device.

[0212] "Means for training generative AI" refers to methods and techniques for collecting data on court records and legal actions and using them to train generative AI models.

[0213] "Means for inputting information to request defense from the generated AI" refers to an interface through which a user can input the information necessary to request defense from the generated AI.

[0214] "Means for generative AI to conduct legal advocacy in court" refers to technologies and systems that enable generative AI to conduct legal advocacy in court based on data learned by the AI ​​and provide legal advice in real time.

[0215] "Means for visually displaying through a user interface" refers to an interface that allows a user to visually check the results of the lawsuit's winning probability calculation and details of similar precedents.

[0216] "Means for providing advice to help users decide whether to proceed with litigation" refers to a system or method for providing information or advice to users that will help them decide whether to proceed with litigation.

[0217] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data for civil and family law cases, and then uses generative AI to conduct defense activities in court. This system operates in cooperation with a server, a terminal, and a user.

[0218] Data collection and storage

[0219] First, the server periodically collects case data on past civil and family law cases from public institutions and legal databases. This data includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties involved. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[0220] Entering lawsuit information and calculating win rate

[0221] Users use a device (such as a smartphone) to enter information related to their legal case. This information includes the specific details of the case, the desired outcome, and supporting documents. The entered data is sent from the device to a server. The server then searches a database based on the user's legal case information and extracts similar past cases. This process uses natural language processing (NLP) algorithms and similarity calculations. Based on the analysis of the extracted similar cases, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the case. The calculation results are generated in the form of a report and sent to the device.

[0222] Generative AI advocacy

[0223] When legal amendments are made, the server collects data on court records and lawyers' legal activities and trains the generated AI. This training data includes minutes, evidentiary documents, and legal content. The generated AI uses the trained data to acquire legal skills to support litigation. When a user requests the generated AI to act as a legal representative, they enter the necessary information on their device and send a request for legal representation to the server. The server then formulates the optimal legal strategy based on this information and the training data. When the generated AI acts as a legal representative in court, it sends advice to the user in real time to support the progress of the trial. The user can check the generated AI's legal activities in court and issue instructions as necessary. Once the outcome of the lawsuit is known, a detailed report is provided via the device, along with advice on future countermeasures.

[0224] User Interface

[0225] The calculation results of the winning rate and details of similar cases sent from the server are visually displayed through the user interface, making it easier for users to intuitively understand the lawsuit information. The same interface also provides advice to help users decide whether to proceed with the lawsuit.

[0226] Specific examples

[0227] For example, suppose a user is considering divorce proceedings and is fighting over asset distribution and child custody. The user uses a terminal to input the details of the lawsuit and supporting documents. The server receives this information and searches a database for similar past divorce cases. Based on the search results, it calculates a 70% chance of success and provides a report to the user. If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend the case. The generated AI has learned from past court records and will defend the case in court. Finally, if the generated AI's defense results in a favorable verdict, the user reviews the results and decides what subsequent actions are required.

[0228] Examples of prompts for generative AI models

[0229] If you are considering divorce proceedings and are disputing asset distribution and child custody, please analyze past legal precedents related to this type of case and propose legal action that focuses on the fairness of asset distribution and the best interests of the children.

[0230] The above is a specific embodiment of the present invention.

[0231] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0232] Step 1: Collect and store data

[0233] The server periodically collects case data for past civil and family law cases from legal-related public data sources and open databases. The data input includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties. This data is stored in a database and managed for easy searching and reference.

[0234] Step 2: Enter your case information

[0235] Users use their devices (smartphones) to input specific information about their legal case—the nature of the case, the desired outcome, evidence, etc.—and this input data is sent from the device to the server.

[0236] Step 3: Find similar cases

[0237] The server searches the database based on the user's lawsuit information received in step 2 to extract similar past cases. This uses natural language processing (NLP) algorithms and similarity calculations. Specifically, the entered lawsuit content is tokenized, vectorized using techniques such as TF-IDF, and compared with similar cases in the database.

[0238] Step 4: Calculating the probability of success of your lawsuit

[0239] The server uses a machine learning model (such as logistic regression or random forest) to calculate the success rate of the lawsuit based on the analysis results of similar cases extracted in step 3. To do this, the input data and statistical data of similar cases are plotted and input into the model to predict the success rate. The calculation results are generated in report format and sent to the terminal.

[0240] Step 5: Litigation Support Report

[0241] The terminal receives the report sent from the server in step 4. The report includes details of similar cases along with the calculated win rate. The user can view this information and use it to decide whether to proceed with the lawsuit.

[0242] Step 6: Generative AI learns to advocate

[0243] When legal changes occur, the server collects court records and data on lawyers' legal activities and trains the AI. This training data includes minutes, evidence, and legal content. The AI ​​uses the collected data to acquire legal skills.

[0244] Step 7: Enter your request for legal representation

[0245] When a user requests a generated AI to defend them, they input the necessary information into their device and send the request to the server, including the specific details of the lawsuit and the key points of the defense they are seeking.

[0246] Step 8: Developing a defense strategy

[0247] The server formulates an optimal defense strategy based on the user information received in step 7 and the data learned by the generating AI. The generating AI utilizes the learned data to provide real-time advice to the user for advocacy in court.

[0248] Step 9: Real-time legal support

[0249] As users pursue their cases in court, the generative AI provides real-time support for their defense, providing advice on how to present evidence and what to say. Users can use this advice to make decisions in court.

[0250] Step 10: Reporting the outcome of the case

[0251] Once the case is over, the device receives a detailed report, including the outcome of the case and the effectiveness of the advice provided by the AI. The user can use this information to decide on future actions.

[0252] These are the specific processing steps of the system. Based on the specific software and algorithms used in each step, we can clearly understand the entire process from data input to output.

[0253] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0254] This invention combines a system that calculates the success rate of lawsuits based on past precedent data from civil and family law lawsuits, and conducts legal defense activities in court using generative AI, with an emotion engine that recognizes the user's emotions. This system consists of a server, a terminal, a user, and an emotion engine.

[0255] Data collection and storage

[0256] The server periodically collects past civil and family law case data from legal-related public data sources and open databases. This data is obtained by normalizing it using web scraping and APIs. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[0257] Entering lawsuit information and calculating win rate

[0258] The user uses the terminal to input information related to their legal case (e.g., the lawsuit content, desired outcome, and supporting documents). During the process of inputting this information, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The emotion recognition results are sent to the server along with the input information.

[0259] The server searches the database based on the received user's lawsuit information and sentiment data to extract similar past cases. This is done using natural language processing (NLP) algorithms and similarity calculations. Based on the analysis results of the extracted similar cases and sentiment data, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the lawsuit.

[0260] Calculating winning rates and providing results

[0261] The server uses an emotion engine to take into account the user's recognized emotional state to more accurately calculate the probability of winning the lawsuit. A report is generated containing the calculation results and details of related similar cases. The report includes detailed explanations to ensure transparency of the calculation process and is presented to the user in a manner that minimizes mental burden.

[0262] The terminal displays the submitted report to the user through a user interface. The user can view the win rate, similar cases, and specific advice. Based on the provided report and advice, the user decides whether to proceed with the lawsuit. Feedback from the emotion engine can also be used to help make a lawsuit decision.

[0263] Generative AI advocacy

[0264] When legal changes are made, the server collects data on court records and lawyers' legal activities and trains the generative AI on this data. This data includes minutes, evidence, and legal content, and is compiled as a training dataset. The generative AI uses the learned data to acquire legal skills to support litigation.

[0265] When a user requests the AI ​​to represent them, they re-enter the necessary legal information into their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the AI. The server then formulates the optimal defense strategy based on the received information and learning data.

[0266] Once litigation begins, the generative AI will actually conduct the defense in court, providing real-time advice to the user and supporting the trial, including suggesting when to present evidence and what questions to ask. Using an emotion engine, the generative AI will continuously monitor the user's emotional state and adjust its strategy as needed.

[0267] Specific examples

[0268] For example, suppose a user is considering divorce proceedings and is battling over asset distribution and child custody. The user uses a device to input the details of the lawsuit and supporting documents. At this time, the emotion engine recognizes emotions such as stress and anxiety from the user's facial expressions and voice. The server receives this information and searches its database for similar past divorce lawsuit cases. Based on the search results, taking emotions into account, the server calculates a 70% chance of success and provides a report to the user.

[0269] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the Generator AI defend them. The emotion engine again recognizes the user's emotions and reflects them in the Generator AI. The Generator AI then conducts the defense in court and provides real-time advice to the user. Finally, if the Generator AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0270] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment where ordinary households can easily receive legal support. Furthermore, by providing support that takes the user's emotions into consideration, it is possible to make more appropriate legal decisions while reducing the mental burden.

[0271] The processing flow will be explained below.

[0272] Step 1:

[0273] The server collects historical civil and family law case data from legal public data sources and open databases, using web scraping and APIs to normalize and retrieve the data.

[0274] Step 2:

[0275] The server preprocesses the collected case data, converting it into a standard format before storing it in a database, including deduplication, filling in missing values, and text cleansing.

[0276] Step 3:

[0277] Users use a terminal to input information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents). During this input process, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state.

[0278] Step 4:

[0279] The terminal transmits the entered lawsuit information and the user's emotion data recognized by the emotion engine to the server. The transmitted data is encoded and sent using a secure communication method.

[0280] Step 5:

[0281] The server analyzes the received user's lawsuit information and sentiment data and searches for similar past cases in its database, using natural language processing (NLP) algorithms and similarity calculations.

[0282] Step 6:

[0283] The server calculates the probability of winning a lawsuit based on the data of similar cases it has searched for, using machine learning models (such as logistic regression or random forest) and taking into account necessary variables.

[0284] Step 7:

[0285] The server adjusts the probability of winning a lawsuit by taking into account the user's emotional state, which is recognized using an emotion engine. For example, if the user is in a high stress state, it is taken into account as a risk factor.

[0286] Step 8:

[0287] The server generates a report detailing the winning probability calculation and related similar cases, including detailed explanations to ensure transparency of the calculation process.

[0288] Step 9:

[0289] The server sends the generated reports to the terminal, which are presented in a visually easy-to-understand format (graphs, tables, text).

[0290] Step 10:

[0291] The terminal displays the submitted report to the user through a user interface, where the user can view the win rate, similar cases, and specific advice.

[0292] Step 11:

[0293] Users can decide whether to proceed with a lawsuit based on the reports and advice provided, and feedback from an emotion engine can also be used to help with the lawsuit decision.

[0294] Step 12:

[0295] When legal changes are made, the server collects data on past court records and lawyers' legal activities, including minutes, evidence, and content of arguments, and uses this data as training data for the generative AI.

[0296] Step 13:

[0297] The server trains the generative AI using data based on court records, using deep learning techniques to train the AI ​​to acquire effective advocacy skills.

[0298] Step 14:

[0299] To request the AI ​​to defend the case, the user re-enters the necessary legal information into the device and submits the request. At this time, the emotion engine again recognizes the user's emotions and reflects them in the AI.

[0300] Step 15:

[0301] The server receives a request for legal representation from the user and formulates an optimal legal representation strategy based on the learning data, and if necessary, communicates with the user.

[0302] Step 16:

[0303] Once litigation begins, the generative AI will actually act as a defense in court, providing real-time advice to the user and supporting the proceedings, including suggesting when to present evidence and what questions to ask.

[0304] Step 17:

[0305] During courtroom advocacy, the generative AI uses an emotion engine to continuously monitor the user's emotional state and adjust its strategy as needed.

[0306] Step 18:

[0307] After the verdict is reached, the server analyzes the results and generates a detailed report that is sent to the device, including the outcome of the verdict and advice on future actions.

[0308] Step 19:

[0309] The user will use the provided report to determine any necessary follow-up actions (eg, appeal decisions, etc.).

[0310] The above are the specific processing steps of the entire system.

[0311] Example 2

[0312] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0313] Conventional litigation support systems could calculate the probability of winning a lawsuit based on past case data, but they were unable to take the user's emotional state into account, resulting in low accuracy in the calculation of the win rate. Furthermore, they were limited in their use of court records and data on lawyers' legal activities, resulting in insufficient courtroom advocacy by the generated AI. Furthermore, they required time and effort to input legal information, and lacked a means to visually present detailed case information. This made it difficult for users to receive appropriate legal support and placed a significant mental burden on them.

[0314] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0315] In this invention, the server includes means for collecting case data on past civil and family law lawsuits and storing it in a database, means for receiving lawsuit information entered by a user and searching the database for similar past cases, means for calculating the success rate of the lawsuit based on the searched similar cases and providing the result to the user, means for analyzing the user's facial expressions and voice and recognizing their emotional state, and means for calculating the success rate by reflecting the emotional state in the calculation means. This enables highly accurate calculation of the success rate that takes the user's emotional state into account, enabling appropriate defense activities in court using generative AI, and providing comprehensive legal support and reducing the mental burden for the user.

[0316] "Past civil and family law case data" refers to a collection of past court decisions and related documents relating to civil and family law cases.

[0317] A "database" is a data management system that systematically stores collected case law data and related information and allows for searching and inquiry.

[0318] "User" refers to an individual or company that utilizes the system to enter information about a lawsuit and receive assistance.

[0319] "Lawsuit information" is information related to the lawsuit entered by the user, including the details of the lawsuit, the desired outcome, and supporting documents.

[0320] "Similar past cases" refers to past precedents that have elements or outcomes similar to the current litigation case.

[0321] "Win rate" is the percentage of the likelihood of success in a lawsuit, calculated based on past similar cases and current litigation information.

[0322] "Analyzing facial expressions and voice to recognize emotional states" refers to the process of reading the user's emotions at that time by analyzing their facial expressions and tone of voice.

[0323] "Generative AI" refers to an artificial intelligence model that learns from past data and automatically responds and makes suggestions to new situations.

[0324] "Courtroom advocacy" refers to a series of activities carried out by lawyers and generative AI in court, such as advocacy, advice, presenting evidence, and proposing questions.

[0325] A "user interface" is a screen or mechanism that allows a user to operate a system, input information, and view results.

[0326] This invention combines a system that calculates the success rate of lawsuits based on past precedent data from civil and family law lawsuits, and conducts legal defense activities in court using generative AI, with an emotion engine that recognizes the user's emotions. This system consists of a server, a terminal, a user, and an emotion engine.

[0327] Data collection and storage

[0328] The server periodically collects past civil and family law case data from legal-related public data sources and open databases. To do this, it uses web scraping tools (e.g., Beautiful Soup, Scrapy) and APIs. The collected data is normalized and checked for quality before being stored in a database (e.g., MySQL, PostgreSQL, MongoDB). The data is managed by removing duplicate data and filling in missing values, and an index is created to enable search and reference as needed.

[0329] Litigation information entry and analysis

[0330] Users use a device (e.g., PC, tablet, or smartphone) to enter information related to their legal case. Specifically, they fill out a form on the device, including the nature of the case, the desired outcome, and supporting documents. During this process, the emotion engine uses the device's camera and microphone to analyze the user's facial expressions and voice to recognize their emotional state in real time. Emotional states include stress levels, anxiety, and joy.

[0331] Submitting Litigation Information

[0332] The terminal transmits the entered litigation information and the emotion data recognized by the emotion engine to the server. This transmission is performed using encrypted communication (e.g., HTTPS).

[0333] Searching for similar cases and calculating the winning rate

[0334] The server searches for similar past cases in its database based on the received user's lawsuit information and emotional data. To do this, it uses natural language processing (NLP) algorithms (e.g., BERT, Word2Vec) and similarity calculation algorithms (e.g., Cosine Similarity). Based on the data of similar cases found, it uses machine learning models (e.g., logistic regression, random forest) to calculate the success rate of the lawsuit. The user's emotional state is also reflected in the prediction model to improve the accuracy of the success rate.

[0335] Generate and provide a report of the results

[0336] The server automatically generates a report containing the winning probability calculation results and details of related similar cases. This report includes detailed explanations to maintain transparency of the calculation process and also serves the purpose of reducing the mental burden on the user. The generated report is displayed to the user via the terminal, and the user can operate it intuitively using the user interface.

[0337] Generative AI advocacy

[0338] When legal amendments are made, the server collects data on court records and lawyers' legal activities and trains this data on a generative AI model (e.g., GPT-3 (registered trademark), BERT). This allows the generative AI to acquire legal skills to support litigation. When a user requests the generative AI to act as a lawyer, they use their device to re-enter the necessary legal information and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the generative AI. The server then formulates an optimal legal strategy based on the received information and learning data. Once the lawsuit begins, the generative AI actually acts as a lawyer in court, providing real-time advice to the user and supporting the trial progress.

[0339] Specific examples

[0340] For example, suppose a user is considering divorce proceedings and is battling over asset distribution and child custody. The user uses a device to input the details of the lawsuit and supporting documents. At this time, the emotion engine recognizes emotions such as stress and anxiety from the user's facial expressions and voice. The server receives this information and searches its database for similar past divorce lawsuit cases. Based on the search results, taking emotions into account, the server calculates a 70% chance of success and provides a report to the user.

[0341] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the Generator AI defend them. The emotion engine again recognizes the user's emotions and reflects them in the Generator AI. The Generator AI then conducts the defense in court and provides real-time advice to the user. Finally, if the Generator AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0342] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment where ordinary households can easily receive legal support. Furthermore, by providing support that takes the user's emotions into consideration, it is possible to make more appropriate legal decisions while reducing the mental burden.

[0343] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0344] Step 1: Data collection and database update

[0345] The server collects past civil and family law case data from legal-related public data sources and open databases. Specifically, data is periodically retrieved using web scraping tools (e.g., Beautiful Soup, Scrapy) and APIs. The collected data is then normalized and its format is checked for quality. This step also involves removing duplicate data and imputing missing values. The input data is obtained from public data sources, and the output data is normalized case data. The data is then stored in a database (e.g., MySQL, PostgreSQL, MongoDB). An index is built for the stored data to enable efficient access for later search and analysis.

[0346] Step 2: Enter case information and recognize emotional state

[0347] Users use a device (e.g., PC, tablet, or smartphone) to enter information related to their legal case. Specifically, they fill out an input form with the details of the lawsuit, the desired outcome, and supporting documents. At the same time, the emotion engine uses the device's camera and microphone to analyze the user's facial expressions and voice to recognize their emotional state in real time. Emotional states include stress levels, anxiety, and joy. The input data are the lawsuit information and emotion data provided by the user, and the output data are the analyzed lawsuit information and emotion state.

[0348] Step 3: Submit your case information

[0349] The terminal transmits the entered legal information and the emotion data recognized by the emotion engine to the server. This transmission is performed using encrypted communication (e.g., HTTPS). The input data is the legal information and emotion data, and the output data is the data securely transmitted to the server.

[0350] Step 4: Find similar cases

[0351] The server searches for similar past cases in the database based on the received user's lawsuit information and emotion data. To do this, it uses natural language processing (NLP) algorithms (e.g., BERT, Word2Vec) and similarity calculation algorithms (e.g., Cosine Similarity). The input data are the received lawsuit information and emotion data, and the output data are the searched similar cases.

[0352] Step 5: Calculating Win Rate

[0353] The server uses data extracted from similar cases to calculate the probability of winning a lawsuit using a machine learning model (e.g., logistic regression, random forest). This step also takes into account the user's emotional data. For example, if the user is very nervous, the predictive model will compensate for that effect. The input data are similar cases and emotional data, and the output data is the calculated probability of winning a lawsuit.

[0354] Step 6: Generate reports

[0355] The server automatically generates a report containing the winning probability calculation results and details of related similar cases. This report includes detailed explanations to maintain transparency of the calculation process and also serves the purpose of reducing the mental burden on the user. The input data are the winning probability calculation results and detailed data of similar cases, and the output data is the generated report.

[0356] Step 7: Provide a report

[0357] The terminal displays the generated report to the user. The user interface is designed to be intuitive. The user views the report through the terminal and decides whether to proceed with the lawsuit. Feedback from the emotion engine is also provided here. The input data is the generated report, and the output data is the report presentation and feedback to the user.

[0358] Step 8: Litigation training of generative AI

[0359] When a legal amendment is made, the server collects data on court records and lawyers' legal activities and trains this data into a generative AI model (e.g., GPT-3, BERT). The input data is the revised court records and legal activities data, and the output data is the model data trained by the generative AI.

[0360] Step 9: Defending the generative AI

[0361] When a user requests a defense from the generative AI, they re-enter the necessary legal information into their device and send the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the generative AI. The server then formulates an optimal defense strategy based on the received information and learning data. The input data is the re-entered legal information and emotion data, and the output data is the defense strategy created by the generative AI.

[0362] Step 10: Real-time advocacy

[0363] Once the lawsuit begins, the generative AI will actually conduct the defense in court. It will provide real-time advice to the user and support the trial, including suggesting when to present evidence and what questions to ask. It will continuously monitor the user's emotional state using an emotion engine, and the generative AI will adjust its strategy as needed. The input data is the user's real-time emotional state and the status of the trial, and the output data is real-time advice and support from the generative AI.

[0364] (Application example 2)

[0365] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0366] While conventional litigation support systems can calculate the probability of winning based on past case data, they do not take into account the user's emotional state and lack support for mental stress and anxiety. Furthermore, because they do not provide user support or emotional monitoring in real time during legal proceedings, they do not alleviate the psychological burden on users in the courtroom. This hinders user decision-making and effective legal proceedings during legal proceedings.

[0367] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0368] In this invention, the server includes means for collecting and storing case data from past civil and family law cases in a database; means for receiving case information entered by a user and searching the database for similar past cases; means for calculating the success rate of the case based on the searched similar cases and providing the result to the user; means for recognizing the user's emotional state using an emotion engine and more accurately calculating the success rate of the case taking the emotional state into account; and means for generating and providing a report that takes the user's emotions into account. This enables accurate calculation of the success rate of the case taking the user's emotional state into account and also provides psychological support for making litigation decisions. Furthermore, by monitoring the user's emotions and providing mental support during real-time legal proceedings, the server can reduce the user's psychological burden and support the effective progress of the case.

[0369] A "civil lawsuit" is a lawsuit to resolve legal disputes between private citizens, where individuals or entities dispute each other's rights and obligations.

[0370] A "family litigation" is a judicial proceeding to resolve disputes about domestic issues and relationships.

[0371] "Case law data" is a record of past decisions made by courts on legal issues.

[0372] A "database" is a system for efficiently searching, storing, and managing large amounts of organized data.

[0373] "Similar cases" are past court decisions that have similar facts and legal issues to those in the current lawsuit.

[0374] A "success rate" is the percentage of chances of success in a lawsuit.

[0375] "Generative AI" is artificial intelligence that generates and analyzes new information based on machine learning and deep learning.

[0376] An "emotion engine" is a system that recognizes and analyzes human emotions from voice, facial expressions, context, etc.

[0377] A "report" is a report that summarizes analysis results and information.

[0378] A "trial record" is a record of all documents and evidence arising in the course of a trial.

[0379] "Advocacy" refers to the work and actions of a lawyer to assist and defend a client under the law.

[0380] A "user interface" is a screen or operating means that allows a user to interact with a system or application.

[0381] "Mental support" refers to services that help users to alleviate and support their mental and emotional anxieties and difficulties.

[0382] This invention is a system that calculates the success rate of lawsuits based on past legal precedent data, and combines advocacy activities using generative AI with user support using an emotion engine. This system consists of a server, terminals, users, and an emotion engine.

[0383] Data collection and storage

[0384] The server collects historical civil and family law case data from legal-related public data sources and open databases using web scraping and APIs, normalizes the data, and stores it in a database that is managed for efficient searching and referencing.

[0385] Entering lawsuit information and calculating win rate

[0386] Using the terminal, the user inputs information related to their legal case. During this information input process, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The emotional state and legal case information are then sent to the server.

[0387] The server searches a database based on the received lawsuit information and sentiment data to extract similar past lawsuits. This search uses natural language processing (NLP) algorithms and similarity calculations. Using the analysis results of the extracted similar cases and sentiment data, a machine learning model (e.g., logistic regression or random forest) is used to calculate the winning rate.

[0388] Calculating winning rates and providing results

[0389] The server uses the emotion engine to take into account the user's emotional state and more accurately calculate the success rate of the lawsuit. A report is generated containing the calculation results and details of related similar cases. The report is displayed to the user through the user interface on their device, allowing them to view the success rate, similar cases, and specific advice. Based on the report and advice provided, the user decides whether to proceed with the lawsuit. Feedback from the emotion engine can also be used to help make legal decisions.

[0390] Generative AI advocacy

[0391] When legal amendments are made or new case data is added, the server collects court records and data on lawyers' legal activities and trains the generative AI. Based on the learned data, the generative AI acquires legal skills to support litigation.

[0392] When a user requests the Generative AI to defend their case, they re-enter the necessary legal information into their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects that information in the Generative AI. The server then formulates an optimal defense strategy based on the received information and learning data. Once the lawsuit begins, the Generative AI will conduct defense activities in court and provide advice to the user in real time. The emotion engine continuously monitors the user's emotional state, and the Generative AI will adjust its strategy as needed.

[0393] Specific examples

[0394] For example, if a company reports a data breach incident, they can use the AI ​​security assistant application to input details of the incident. The emotion engine analyzes the stress and anxiety of the person in charge and compares it with similar cases in the database to conduct a risk assessment. The results are provided as a report, and any necessary mental support is also provided.

[0395] Example prompt sentence:

[0396] "Important data was leaked by an internal employee. As a countermeasure, we are considering strengthening log management and providing employee training."

[0397] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0398] Step 1:

[0399] The server collects past civil and family law case data from legal public data sources and open databases, using web scraping and APIs, and stores it in a database. The data is then normalized and stored in the database.

[0400] Input: Public legal data sources

[0401] Output: Normalized case data

[0402] Step 2:

[0403] The user uses the device to input information related to their legal case (e.g., the details of the case, the desired outcome, and supporting documents). As the user inputs information, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state.

[0404] Input: Lawsuit case information, user facial expression and voice data

[0405] Output: Input case information, perceived emotional state

[0406] Step 3:

[0407] The server searches the database based on the received user's lawsuit information and emotion data to extract similar past lawsuits, using natural language processing (NLP) algorithms and similarity calculations.

[0408] Input: User's lawsuit information, emotional data

[0409] Output: Extracted similar lawsuit cases

[0410] Step 4:

[0411] Based on the extracted similar lawsuits and sentiment data, the server uses machine learning models to calculate the probability of winning a lawsuit, using algorithms such as logistic regression and random forests.

[0412] Input: Similar lawsuit cases, emotional data

[0413] Output: Calculated success rate of lawsuits

[0414] Step 5:

[0415] The server further adjusts the chances of winning the case by taking into account the emotional state of the user as recognized using the emotion engine.

[0416] Inputs: Calculated case win rate, perceived emotional state

[0417] Output: Adjusted lawsuit win rate

[0418] Step 6:

[0419] The server generates a report containing the results of the calculation and details of related similar cases, which can be transmitted to the terminal and viewed by the user through the user interface.

[0420] Input: Adjusted case win rate, details of related similar cases

[0421] Output: Generated report

[0422] Step 7:

[0423] The user decides whether to proceed with the lawsuit based on the report and advice provided through the terminal. During this decision-making process, the emotion engine monitors the user's emotional state and provides feedback.

[0424] Input: Report, Advice

[0425] Output: User's legal decision

[0426] Step 8:

[0427] When legal amendments or new case law data are added, the server collects court records and data on lawyers' legal activities and trains the generation AI.

[0428] Input: Latest court records and advocacy data

[0429] Output: A trained generative AI model

[0430] Step 9:

[0431] When a user requests the AI ​​to represent them, they re-enter the case information into their device, which is then sent to the server. During this process, the emotion engine recognizes the user's emotions and reflects them in the AI.

[0432] Input: Re-entered case information, perceived emotional state

[0433] Output: Data passed to the generative AI

[0434] Step 10:

[0435] The server uses the trained generative AI model to develop an optimal defense strategy and provides real-time advice to the user during the courtroom. The emotion engine continuously monitors the user's emotional state, and the generative AI adjusts the strategy.

[0436] Input: Generative AI model, real-time user emotion data

[0437] Output: Real-time advice and tailored defense strategies

[0438] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0439] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0440] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0441] [Second embodiment]

[0442] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0443] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0444] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0445] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0446] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0447] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0448] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0449] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0450] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0451] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0452] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0453] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0454] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data of civil and family law lawsuits, and furthermore, uses generative AI to conduct defense activities in court. This system operates in cooperation with three parties: a server, a terminal, and a user.

[0455] Data collection and storage

[0456] The server periodically collects case data for past civil and family law cases from legal-related public data sources and open databases. This data includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[0457] Entering lawsuit information and calculating win rate

[0458] The user uses the terminal to input information related to their legal case, including the specifics of the case, the desired outcome, and supporting documents. This input data is then sent from the terminal to the server.

[0459] The server searches the database based on the received user's lawsuit information to extract similar past cases. This is done using natural language processing (NLP) algorithms and similarity calculations. Based on the analysis results of the extracted similar cases, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the lawsuit. The calculation results are generated in report format and sent to the terminal.

[0460] The user can review the reports provided by the terminal and view details of the winning rate and related cases. Based on this information, the user can decide whether to proceed with the lawsuit.

[0461] Generative AI advocacy

[0462] When legal changes are made, the server collects data on court records and lawyers' legal activities and trains the generative AI. This training data includes minutes, evidence, and legal content. The generative AI acquires legal skills to support litigation based on the learned data.

[0463] When a user requests that the generated AI represent them, they input the necessary information into their device and send a request for the generated AI to the server. The server then uses this information and learning data to formulate an optimal defense strategy. When the generated AI represents its case in court, it provides real-time advice to the user and supports the progress of the trial.

[0464] Users can monitor the AI's defense activities in court and provide instructions as necessary. Once the outcome of the lawsuit is known, a detailed report will be provided via the device, along with advice on future actions.

[0465] Specific examples

[0466] For example, suppose a user is considering divorce proceedings and is fighting over asset division and child custody. The user uses a terminal to enter the details of the lawsuit and supporting documents. The server receives this information and searches its database for similar past divorce cases. Based on the search results, the server calculates a 70% chance of success and provides a report to the user.

[0467] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend their case. The generated AI will study past court records and conduct a defense in court. Finally, if the generated AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0468] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment in which ordinary households can easily receive legal support.

[0469] The processing flow will be explained below.

[0470] Step 1:

[0471] The server collects historical civil and family law case data from legal public data sources and open databases, using web scraping and APIs to normalize and retrieve the data.

[0472] Step 2:

[0473] The server preprocesses the collected case data, converting it into a standard format before storing it in a database, including deduplication, filling in missing values, and text cleansing.

[0474] Step 3:

[0475] The user uses a terminal to enter information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents). The input form has required and optional fields, and the user must enter all required information.

[0476] Step 4:

[0477] The terminal transmits the entered litigation information to the server, where the transmitted data is encoded and sent using a secure communication method.

[0478] Step 5:

[0479] The server analyzes the received user's lawsuit information and searches for similar past cases in its database, using natural language processing (NLP) algorithms and similarity calculations.

[0480] Step 6:

[0481] The server calculates the probability of winning a lawsuit based on the data of similar cases retrieved, using machine learning models (such as logistic regression and random forests) to take into account necessary variables.

[0482] Step 7:

[0483] The server generates a report detailing the winning probability calculation and related similar cases, including a detailed explanation of the calculation process to ensure transparency.

[0484] Step 8:

[0485] The server sends the generated reports to the terminal, which are presented in a visually easy-to-understand format (graphs, tables, text).

[0486] Step 9:

[0487] The terminal displays the submitted report to the user through a user interface, where the user can view the win rate, similar cases, and specific advice.

[0488] Step 10:

[0489] Users can decide whether to proceed with a lawsuit based on the reports and advice provided, and can also ask additional questions or complete the data.

[0490] Step 11:

[0491] When legal changes are made, the server collects data on court records and lawyers' legal activities, including minutes, evidence, and content of legal arguments, and organizes it into a learning dataset.

[0492] Step 12:

[0493] The server trains the generative AI using data based on court records, using deep learning techniques to train the AI ​​to acquire effective advocacy skills.

[0494] Step 13:

[0495] To request the generated AI to defend the case, the user re-enters the necessary legal information into the device and submits the request.

[0496] Step 14:

[0497] The server receives a request for legal representation from the user and formulates an optimal legal representation strategy based on the learning data, and if necessary, communicates with the user.

[0498] Step 15:

[0499] Once litigation begins, the generative AI will actually act as a defense in court, providing real-time advice to the user and supporting the proceedings, including suggesting when to present evidence and what questions to ask.

[0500] Step 16:

[0501] After the verdict is reached, the server analyzes the results and generates a detailed report that is sent to the device, including the outcome of the verdict and advice on future actions.

[0502] Step 17:

[0503] The user will use the provided report to determine any necessary follow-up actions (eg, appeal decisions, etc.).

[0504] The above are the specific processing steps of the entire system.

[0505] Example 1

[0506] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0507] Existing legal support systems have difficulty calculating an appropriate win rate for each user's legal information, or providing real-time support for legal activities based on past cases. Furthermore, to efficiently support legal activities in court, it is necessary to analyze a huge amount of past court records and lawyers' legal activity data, and use generative AI to formulate optimal legal strategies.

[0508] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0509] In this invention, the server includes means for collecting past litigation case data from legal-related public data sources and open databases and storing it in a database, means for receiving litigation information entered by a user and searching the database for similar past cases using a natural language processing algorithm, means for analyzing the searched similar cases and calculating the success rate of the litigation using a machine learning model and providing the results to the user in the form of a report, and means for the user to use a terminal for inputting litigation information and evidential materials. This allows users to calculate the success rate of their own litigation cases with high accuracy and to develop optimal defense strategies based on similar past cases.

[0510] "Legal-related public data sources" refers to databases of laws, precedents, court records, etc. that are publicly available from governments and related agencies.

[0511] "Open database" refers to a database provided by a third party that is freely accessible on the Internet.

[0512] "Litigation case data" refers to a collection of data including records of past civil and family law litigation, judgments, related laws and regulations, evidence lists, and background information on the parties involved.

[0513] "Database" refers to an information system for efficiently storing, retrieving, and managing information.

[0514] "Law case information entered by the user" refers to the specific details, desired outcome, supporting documents, etc., of the user's own legal case.

[0515] "Natural language processing algorithm" refers to technology that allows computers to analyze human language and understand its meaning.

[0516] "Similar past cases" refers to past legal cases that have similar characteristics based on specific criteria to the user's legal information.

[0517] A "machine learning model" refers to a statistical algorithm that uses data to learn and make predictions, classifications, and other similar tasks.

[0518] "Generative AI" refers to artificial intelligence that can learn from past data and automatically perform specific tasks.

[0519] A "defense strategy" refers to a plan or method that aims to gain an advantage in a lawsuit based on law and evidence.

[0520] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data for civil and family law lawsuits, and uses generative AI to conduct defense activities in court. This system operates in cooperation with three parties: a server, a terminal, and a user.

[0521] Data collection and storage

[0522] The server periodically collects past civil and family law case data from legal-related public data sources and open databases using curl or API. The collected data is then stored in a database (e.g., MySQL or PostgreSQL) after checking its accuracy. It is also indexed to streamline data search and management.

[0523] Entering lawsuit information and calculating win rate

[0524] The user enters information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents) into a dedicated form on the device. The device temporarily stores the entered data, and once all the data is collected, it sends an HTTP POST request to the server.

[0525] The server uses a natural language processing (NLP) algorithm to search its database for similar past cases based on the received user's lawsuit information. Specifically, it generates a search query and performs a full-text search of the past lawsuit database. Based on the analysis results of the extracted similar cases, it uses a machine learning model (e.g., logistic regression or random forest) to calculate the success rate of the lawsuit.

[0526] The results are formatted into a report and sent from the server to the terminal. The user can then review the report provided by the terminal and refer to the winning rate and details of related cases. Based on this information, the user can decide whether to proceed with the lawsuit.

[0527] Generative AI advocacy

[0528] When legal changes or new legal precedents arise, the server collects court records and lawyers' advocacy data and trains the generative AI using a natural language processing framework (e.g., BERT). This training data includes court records, evidence, and advocacy content.

[0529] When a user requests a defense from the generated AI, they input the necessary information into their device and send a defense request to the server. The server receives this information and provides the necessary data to the generated AI.

[0530] The generative AI will develop an optimal defense strategy based on past learning data and support legal actions in court. It will send advice to the user in real time to support the trial. The user can monitor the generative AI's legal actions in court and give instructions as necessary. Once the outcome of the trial is known, a detailed report will be provided to the user via their device, along with advice on future countermeasures.

[0531] Specific examples

[0532] For example, suppose a user is considering divorce proceedings and is fighting over asset division and child custody. The user uses a terminal to enter the details of the lawsuit and supporting documents. The server receives this information and searches its database for similar past divorce cases. Based on the search results, the server calculates a 70% chance of success and provides a report to the user.

[0533] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend their case. The generated AI will learn from past court records and conduct a defense in court. Finally, if the generated AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0534] An example prompt is, "I am currently in a divorce proceeding regarding asset division and child custody. How can I calculate my chances of winning based on past case data and use generative AI to defend my case in court?"

[0535] The above is a specific embodiment of the present invention. This system will reduce the cost and time required for litigation, and create an environment in which ordinary households can easily receive legal support.

[0536] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0537] Step 1:

[0538] The server periodically collects past civil and family law case data from legal-related public data sources and open databases via the Internet, using curl and APIs to retrieve the data.

[0539] Input: Data source URL or API endpoint

[0540] Data processing: Formatting the acquired data and checking its accuracy

[0541] Output: A formatted dataset

[0542] Specific example: "curl https: / / example-legal-database.com / api / cases"

[0543] Step 2:

[0544] The server stores the collected data in a database (e.g., MySQL or PostgreSQL) and indexes it for efficient searching.

[0545] Input: Formatted dataset

[0546] Data manipulation: Inserting and indexing data based on the database schema

[0547] Output: Data stored in the database

[0548] Specific example: "INSERT INTO cases_table (case_id, case_type, verdict, laws, evidence, parties_info) VALUES (...)"

[0549] Step 3:

[0550] The user enters the case information into a dedicated form on the terminal.

[0551] Input: Contents of lawsuit, desired outcome, evidence, etc.

[0552] Data calculation: Temporarily save input data

[0553] Output: Temporarily saved dataset

[0554] Specific example: "Enter the details of the lawsuit in text on a web form"

[0555] Step 4:

[0556] The terminal sends the completed input data to the server via an HTTP POST request.

[0557] Input: Completed dataset

[0558] Data processing: Generating the request body

[0559] Output: Request sent to server

[0560] Example: "POST / api / case_submission { case_data: {...}}"

[0561] Step 5:

[0562] Based on the user's lawsuit information received by the server, a natural language processing (NLP) algorithm is used to search for similar past cases.

[0563] Input: User's case information

[0564] Data Computing: Database Querying Using Full-Text Search Algorithms

[0565] Output: List of similar jobs

[0566] Specific example: SELECT FROM cases_table WHERE MATCH (case_text) AGAINST ('User's lawsuit details')

[0567] Step 6:

[0568] The server analyzes data from similar cases and uses machine learning models (such as logistic regression or random forest) to calculate the probability of winning a lawsuit.

[0569] Input: Data for similar cases

[0570] Data Computation: Probability Computation Using Machine Learning Models

[0571] Output: Win rate calculation result

[0572] Specific example: "Calculating probability using a machine learning model (e.g., sklearn's LogisticRegression.predict_proba)"

[0573] Step 7:

[0574] The server formats the winning rate calculation results into a report format and sends it to the terminal.

[0575] Input: Win rate calculation result

[0576] Data processing: Report generation (e.g. HTML or PDF format)

[0577] Output: Report sent to user terminal

[0578] Specific example: "Using a report generation library (e.g. ReportLab)"

[0579] Step 8:

[0580] The user reviews the report on the terminal and decides whether to proceed with the lawsuit.

[0581] Input: Report data

[0582] Data calculation: Understanding and evaluating report content

[0583] Output: Decision on next action

[0584] Specific example: "Display report PDF on browser"

[0585] Step 9:

[0586] The user inputs the necessary information to request legal representation from the generated AI and sends it from the terminal to the server.

[0587] Input: Detailed information about the legal request

[0588] Data processing: Generating request data

[0589] Output: Request sent to server

[0590] Specific example: "Enter legal request information on a web form"

[0591] Step 10:

[0592] The server provides the received request information to the generation AI, which then formulates the optimal defense strategy.

[0593] Input: Legal request information

[0594] Data Computation: Strategy Planning with Generative AI

[0595] Output: Strategic proposal

[0596] Specific example: "API call to pass request information to the generation AI"

[0597] Step 11:

[0598] Generative AI will act as a defense in court and provide real-time advice to users.

[0599] Input: Strategic data generated by generative AI

[0600] Data Computing: Real-time Advice Generation

[0601] Output: Provide advice to the user

[0602] Specific example of operation: "Generative AI sends advice in real time in chat format"

[0603] Step 12:

[0604] The user will monitor the generated AI's defense activities in court and provide instructions as necessary.

[0605] Input: Generative AI advice and on-site situation

[0606] Data Computing: Advice and Situation Assessment

[0607] Output: Sending instructions

[0608] Specific example: "Real-time feedback and instructions on web applications"

[0609] Step 13:

[0610] When the outcome of the lawsuit becomes clear, the server generates a detailed report and sends it to the device, along with advice on future countermeasures.

[0611] Input: Lawsuit outcome data

[0612] Data processing: detailed report generation and analysis of future countermeasures

[0613] Output: Detailed report and further advice to the user

[0614] Specific example: "Generating and sending a final result report"

[0615] (Application example 1)

[0616] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0617] In the past, in civil and family law cases, it was difficult to calculate the success rate of a case based on past case data, and there was a lack of means to provide real-time advice to properly support legal advocacy in court. As a result, in many cases, users found it difficult to make informed decisions, and the quality of legal advocacy was inconsistent.

[0618] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0619] In this invention, the server includes means for collecting case data on past civil and family law litigation cases and storing it in a database, means for receiving litigation information entered by a user and searching the database for similar past cases, means for calculating the success rate of the litigation based on the searched similar cases and providing the result to the user in real time, and means for receiving the litigation information entered by a user via a smartphone and calculating the success rate of the litigation in real time. This allows the user to obtain a highly accurate litigation success rate based on past cases and to obtain the necessary information in real time via their smartphone.

[0620] "Past civil and family law case precedent data" refers to past civil and family law case precedent information collected from public institutions and legal databases.

[0621] "Means of storing in a database" refers to a method or system for systematically organizing collected case law data and storing it so that it can be easily searched and analyzed.

[0622] "Means for receiving litigation information entered by the user and searching for similar past cases" refers to technologies and algorithms for obtaining information about litigation provided by the user and using that information to find similar cases from past case data.

[0623] "Means for calculating the success rate of a lawsuit and providing the result to the user in real time" refers to technology or methods that analyze past case law data and information provided by the user to calculate the likelihood that the outcome of the lawsuit will be favorable to the user and present the result to the user in real time.

[0624] "Means of obtaining necessary information in real time via a smartphone" refers to applications and systems that allow users to receive information provided in real time via a smartphone, which is a mobile device.

[0625] "Means for training generative AI" refers to methods and techniques for collecting data on court records and legal actions and using them to train generative AI models.

[0626] "Means for inputting information to request defense from the generated AI" refers to an interface through which a user can input the information necessary to request defense from the generated AI.

[0627] "Means for generative AI to conduct legal advocacy in court" refers to technologies and systems that enable generative AI to conduct legal advocacy in court based on data learned by the AI ​​and provide legal advice in real time.

[0628] "Means for visually displaying through a user interface" refers to an interface that allows a user to visually check the results of the lawsuit's winning probability calculation and details of similar precedents.

[0629] "Means for providing advice to help users decide whether to proceed with litigation" refers to a system or method for providing information or advice to users that will help them decide whether to proceed with litigation.

[0630] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data for civil and family law cases, and then uses generative AI to conduct defense activities in court. This system operates in cooperation with a server, a terminal, and a user.

[0631] Data collection and storage

[0632] First, the server periodically collects case data on past civil and family law cases from public institutions and legal databases. This data includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties involved. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[0633] Entering lawsuit information and calculating win rate

[0634] Users use a device (such as a smartphone) to enter information related to their legal case. This information includes the specific details of the case, the desired outcome, and supporting documents. The entered data is sent from the device to a server. The server then searches a database based on the user's legal case information and extracts similar past cases. This process uses natural language processing (NLP) algorithms and similarity calculations. Based on the analysis of the extracted similar cases, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the case. The calculation results are generated in the form of a report and sent to the device.

[0635] Generative AI advocacy

[0636] When legal amendments are made, the server collects data on court records and lawyers' legal activities and trains the generated AI. This training data includes minutes, evidentiary documents, and legal content. The generated AI uses the trained data to acquire legal skills to support litigation. When a user requests the generated AI to act as a legal representative, they enter the necessary information on their device and send a request for legal representation to the server. The server then formulates the optimal legal strategy based on this information and the training data. When the generated AI acts as a legal representative in court, it sends advice to the user in real time to support the progress of the trial. The user can check the generated AI's legal activities in court and issue instructions as necessary. Once the outcome of the lawsuit is known, a detailed report is provided via the device, along with advice on future countermeasures.

[0637] User Interface

[0638] The calculation results of the winning rate and details of similar cases sent from the server are visually displayed through the user interface, making it easier for users to intuitively understand the lawsuit information. The same interface also provides advice to help users decide whether to proceed with the lawsuit.

[0639] Specific examples

[0640] For example, suppose a user is considering divorce proceedings and is fighting over asset distribution and child custody. The user uses a terminal to input the details of the lawsuit and supporting documents. The server receives this information and searches a database for similar past divorce cases. Based on the search results, it calculates a 70% chance of success and provides a report to the user. If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend the case. The generated AI has learned from past court records and will defend the case in court. Finally, if the generated AI's defense results in a favorable verdict, the user reviews the results and decides what subsequent actions are required.

[0641] Examples of prompts for generative AI models

[0642] If you are considering divorce proceedings and are disputing asset distribution and child custody, please analyze past legal precedents related to this type of case and propose legal action that focuses on the fairness of asset distribution and the best interests of the children.

[0643] The above is a specific embodiment of the present invention.

[0644] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0645] Step 1: Collect and store data

[0646] The server periodically collects case data for past civil and family law cases from legal-related public data sources and open databases. The data input includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties. This data is stored in a database and managed for easy searching and reference.

[0647] Step 2: Enter your case information

[0648] Users use their devices (smartphones) to input specific information about their legal case—the nature of the case, the desired outcome, evidence, etc.—and this input data is sent from the device to the server.

[0649] Step 3: Find similar cases

[0650] The server searches the database based on the user's lawsuit information received in step 2 to extract similar past cases. This uses natural language processing (NLP) algorithms and similarity calculations. Specifically, the entered lawsuit content is tokenized, vectorized using techniques such as TF-IDF, and compared with similar cases in the database.

[0651] Step 4: Calculating the probability of success of your lawsuit

[0652] The server uses a machine learning model (such as logistic regression or random forest) to calculate the success rate of the lawsuit based on the analysis results of similar cases extracted in step 3. To do this, the input data and statistical data of similar cases are plotted and input into the model to predict the success rate. The calculation results are generated in report format and sent to the terminal.

[0653] Step 5: Litigation Support Report

[0654] The terminal receives the report sent from the server in step 4. The report includes details of similar cases along with the calculated win rate. The user can view this information and use it to decide whether to proceed with the lawsuit.

[0655] Step 6: Generative AI learns to advocate

[0656] When legal changes occur, the server collects court records and data on lawyers' legal activities and trains the AI. This training data includes minutes, evidence, and legal content. The AI ​​uses the collected data to acquire legal skills.

[0657] Step 7: Enter your request for legal representation

[0658] When a user requests a generated AI to defend them, they input the necessary information into their device and send the request to the server, including the specific details of the lawsuit and the key points of the defense they are seeking.

[0659] Step 8: Developing a defense strategy

[0660] The server formulates an optimal defense strategy based on the user information received in step 7 and the data learned by the generating AI. The generating AI utilizes the learned data to provide real-time advice to the user for advocacy in court.

[0661] Step 9: Real-time legal support

[0662] As users pursue their cases in court, the generative AI provides real-time support for their defense, providing advice on how to present evidence and what to say. Users can use this advice to make decisions in court.

[0663] Step 10: Reporting the outcome of the case

[0664] Once the case is over, the device receives a detailed report, including the outcome of the case and the effectiveness of the advice provided by the AI. The user can use this information to decide on future actions.

[0665] These are the specific processing steps of the system. Based on the specific software and algorithms used in each step, we can clearly understand the entire process from data input to output.

[0666] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0667] This invention combines a system that calculates the success rate of lawsuits based on past precedent data from civil and family law lawsuits, and conducts legal defense activities in court using generative AI, with an emotion engine that recognizes the user's emotions. This system consists of a server, a terminal, a user, and an emotion engine.

[0668] Data collection and storage

[0669] The server periodically collects past civil and family law case data from legal-related public data sources and open databases. This data is obtained by normalizing it using web scraping and APIs. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[0670] Entering lawsuit information and calculating win rate

[0671] The user uses the terminal to input information related to their legal case (e.g., the lawsuit content, desired outcome, and supporting documents). During the process of inputting this information, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The emotion recognition results are sent to the server along with the input information.

[0672] The server searches the database based on the received user's lawsuit information and sentiment data to extract similar past cases. This is done using natural language processing (NLP) algorithms and similarity calculations. Based on the analysis results of the extracted similar cases and sentiment data, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the lawsuit.

[0673] Calculating winning rates and providing results

[0674] The server uses an emotion engine to take into account the user's recognized emotional state to more accurately calculate the probability of winning the lawsuit. A report is generated containing the calculation results and details of related similar cases. The report includes detailed explanations to ensure transparency of the calculation process and is presented to the user in a manner that minimizes mental burden.

[0675] The terminal displays the submitted report to the user through a user interface. The user can view the win rate, similar cases, and specific advice. Based on the provided report and advice, the user decides whether to proceed with the lawsuit. Feedback from the emotion engine can also be used to help make a lawsuit decision.

[0676] Generative AI advocacy

[0677] When legal changes are made, the server collects data on court records and lawyers' legal activities and trains the generative AI on this data. This data includes minutes, evidence, and legal content, and is compiled as a training dataset. The generative AI uses the learned data to acquire legal skills to support litigation.

[0678] When a user requests the AI ​​to represent them, they re-enter the necessary legal information into their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the AI. The server then formulates the optimal defense strategy based on the received information and learning data.

[0679] Once litigation begins, the generative AI will actually conduct the defense in court, providing real-time advice to the user and supporting the trial, including suggesting when to present evidence and what questions to ask. Using an emotion engine, the generative AI will continuously monitor the user's emotional state and adjust its strategy as needed.

[0680] Specific examples

[0681] For example, suppose a user is considering divorce proceedings and is battling over asset distribution and child custody. The user uses a device to input the details of the lawsuit and supporting documents. At this time, the emotion engine recognizes emotions such as stress and anxiety from the user's facial expressions and voice. The server receives this information and searches its database for similar past divorce lawsuit cases. Based on the search results, taking emotions into account, the server calculates a 70% chance of success and provides a report to the user.

[0682] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the Generator AI defend them. The emotion engine again recognizes the user's emotions and reflects them in the Generator AI. The Generator AI then conducts the defense in court and provides real-time advice to the user. Finally, if the Generator AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0683] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment where ordinary households can easily receive legal support. Furthermore, by providing support that takes the user's emotions into consideration, it is possible to make more appropriate legal decisions while reducing the mental burden.

[0684] The processing flow will be explained below.

[0685] Step 1:

[0686] The server collects historical civil and family law case data from legal public data sources and open databases, using web scraping and APIs to normalize and retrieve the data.

[0687] Step 2:

[0688] The server preprocesses the collected case data, converting it into a standard format before storing it in a database, including deduplication, filling in missing values, and text cleansing.

[0689] Step 3:

[0690] Users use a terminal to input information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents). During this input process, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state.

[0691] Step 4:

[0692] The terminal transmits the entered lawsuit information and the user's emotion data recognized by the emotion engine to the server. The transmitted data is encoded and sent using a secure communication method.

[0693] Step 5:

[0694] The server analyzes the received user's lawsuit information and sentiment data and searches for similar past cases in its database, using natural language processing (NLP) algorithms and similarity calculations.

[0695] Step 6:

[0696] The server calculates the probability of winning a lawsuit based on the data of similar cases it has searched for, using machine learning models (such as logistic regression or random forest) and taking into account necessary variables.

[0697] Step 7:

[0698] The server adjusts the probability of winning a lawsuit by taking into account the user's emotional state, which is recognized using an emotion engine. For example, if the user is in a high stress state, it is taken into account as a risk factor.

[0699] Step 8:

[0700] The server generates a report detailing the winning probability calculation and related similar cases, including detailed explanations to ensure transparency of the calculation process.

[0701] Step 9:

[0702] The server sends the generated reports to the terminal, which are presented in a visually easy-to-understand format (graphs, tables, text).

[0703] Step 10:

[0704] The terminal displays the submitted report to the user through a user interface, where the user can view the win rate, similar cases, and specific advice.

[0705] Step 11:

[0706] Users can decide whether to proceed with a lawsuit based on the reports and advice provided, and feedback from an emotion engine can also be used to help with the lawsuit decision.

[0707] Step 12:

[0708] When legal changes are made, the server collects data on past court records and lawyers' legal activities, including minutes, evidence, and content of arguments, and uses this data as training data for the generative AI.

[0709] Step 13:

[0710] The server trains the generative AI using data based on court records, using deep learning techniques to train the AI ​​to acquire effective advocacy skills.

[0711] Step 14:

[0712] To request the AI ​​to defend the case, the user re-enters the necessary legal information into the device and submits the request. At this time, the emotion engine again recognizes the user's emotions and reflects them in the AI.

[0713] Step 15:

[0714] The server receives a request for legal representation from the user and formulates an optimal legal representation strategy based on the learning data, and if necessary, communicates with the user.

[0715] Step 16:

[0716] Once litigation begins, the generative AI will actually act as a defense in court, providing real-time advice to the user and supporting the proceedings, including suggesting when to present evidence and what questions to ask.

[0717] Step 17:

[0718] During courtroom advocacy, the generative AI uses an emotion engine to continuously monitor the user's emotional state and adjust its strategy as needed.

[0719] Step 18:

[0720] After the verdict is reached, the server analyzes the results and generates a detailed report that is sent to the device, including the outcome of the verdict and advice on future actions.

[0721] Step 19:

[0722] The user will use the provided report to determine any necessary follow-up actions (eg, appeal decisions, etc.).

[0723] The above are the specific processing steps of the entire system.

[0724] Example 2

[0725] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0726] Conventional litigation support systems could calculate the probability of winning a lawsuit based on past case data, but they were unable to take the user's emotional state into account, resulting in low accuracy in the calculation of the win rate. Furthermore, they were limited in their use of court records and data on lawyers' legal activities, resulting in insufficient courtroom advocacy by the generated AI. Furthermore, they required time and effort to input legal information, and lacked a means to visually present detailed case information. This made it difficult for users to receive appropriate legal support and placed a significant mental burden on them.

[0727] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0728] In this invention, the server includes means for collecting case data on past civil and family law lawsuits and storing it in a database, means for receiving lawsuit information entered by a user and searching the database for similar past cases, means for calculating the success rate of the lawsuit based on the searched similar cases and providing the result to the user, means for analyzing the user's facial expressions and voice and recognizing their emotional state, and means for calculating the success rate by reflecting the emotional state in the calculation means. This enables highly accurate calculation of the success rate that takes the user's emotional state into account, enabling appropriate defense activities in court using generative AI, and providing comprehensive legal support and reducing the mental burden for the user.

[0729] "Past civil and family law case data" refers to a collection of past court decisions and related documents relating to civil and family law cases.

[0730] A "database" is a data management system that systematically stores collected case law data and related information and allows for searching and inquiry.

[0731] "User" refers to an individual or company that utilizes the system to enter information about a lawsuit and receive assistance.

[0732] "Lawsuit information" is information related to the lawsuit entered by the user, including the details of the lawsuit, the desired outcome, and supporting documents.

[0733] "Similar past cases" refers to past precedents that have elements or outcomes similar to the current litigation case.

[0734] "Win rate" is the percentage of the likelihood of success in a lawsuit, calculated based on past similar cases and current litigation information.

[0735] "Analyzing facial expressions and voice to recognize emotional states" refers to the process of reading the user's emotions at that time by analyzing their facial expressions and tone of voice.

[0736] "Generative AI" refers to an artificial intelligence model that learns from past data and automatically responds and makes suggestions to new situations.

[0737] "Courtroom advocacy" refers to a series of activities carried out by lawyers and generative AI in court, such as advocacy, advice, presenting evidence, and proposing questions.

[0738] A "user interface" is a screen or mechanism that allows a user to operate a system, input information, and view results.

[0739] This invention combines a system that calculates the success rate of lawsuits based on past precedent data from civil and family law lawsuits, and conducts legal defense activities in court using generative AI, with an emotion engine that recognizes the user's emotions. This system consists of a server, a terminal, a user, and an emotion engine.

[0740] Data collection and storage

[0741] The server periodically collects past civil and family law case data from legal-related public data sources and open databases. To do this, it uses web scraping tools (e.g., Beautiful Soup, Scrapy) and APIs. The collected data is normalized and checked for quality before being stored in a database (e.g., MySQL, PostgreSQL, MongoDB). The data is managed by removing duplicate data and filling in missing values, and an index is created to enable search and reference as needed.

[0742] Litigation information entry and analysis

[0743] Users use a device (e.g., PC, tablet, or smartphone) to enter information related to their legal case. Specifically, they fill out a form on the device, including the nature of the case, the desired outcome, and supporting documents. During this process, the emotion engine uses the device's camera and microphone to analyze the user's facial expressions and voice to recognize their emotional state in real time. Emotional states include stress levels, anxiety, and joy.

[0744] Submitting Litigation Information

[0745] The terminal transmits the entered litigation information and the emotion data recognized by the emotion engine to the server. This transmission is performed using encrypted communication (e.g., HTTPS).

[0746] Searching for similar cases and calculating the winning rate

[0747] The server searches for similar past cases in its database based on the received user's lawsuit information and emotional data. To do this, it uses natural language processing (NLP) algorithms (e.g., BERT, Word2Vec) and similarity calculation algorithms (e.g., Cosine Similarity). Based on the data of similar cases found, it uses machine learning models (e.g., logistic regression, random forest) to calculate the success rate of the lawsuit. The user's emotional state is also reflected in the prediction model to improve the accuracy of the success rate.

[0748] Generate and provide a report of the results

[0749] The server automatically generates a report containing the winning probability calculation results and details of related similar cases. This report includes detailed explanations to maintain transparency of the calculation process and also serves the purpose of reducing the mental burden on the user. The generated report is displayed to the user via the terminal, and the user can operate it intuitively using the user interface.

[0750] Generative AI advocacy

[0751] When legal amendments are made, the server collects data on court records and lawyers' legal activities and trains this data on a generative AI model (e.g., GPT-3, BERT). This allows the generative AI to acquire legal skills to support litigation. When a user requests the generative AI to act as a legal representative, they re-enter the necessary legal information using their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the generative AI. The server then formulates an optimal legal strategy based on the received information and learning data. Once the lawsuit begins, the generative AI actually acts as a legal representative in court, providing real-time advice to the user and supporting the trial progress.

[0752] Specific examples

[0753] For example, suppose a user is considering divorce proceedings and is battling over asset distribution and child custody. The user uses a device to input the details of the lawsuit and supporting documents. At this time, the emotion engine recognizes emotions such as stress and anxiety from the user's facial expressions and voice. The server receives this information and searches its database for similar past divorce lawsuit cases. Based on the search results, taking emotions into account, the server calculates a 70% chance of success and provides a report to the user.

[0754] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the Generator AI defend them. The emotion engine again recognizes the user's emotions and reflects them in the Generator AI. The Generator AI then conducts the defense in court and provides real-time advice to the user. Finally, if the Generator AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0755] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment where ordinary households can easily receive legal support. Furthermore, by providing support that takes the user's emotions into consideration, it is possible to make more appropriate legal decisions while reducing the mental burden.

[0756] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0757] Step 1: Data collection and database update

[0758] The server collects past civil and family law case data from legal-related public data sources and open databases. Specifically, data is periodically retrieved using web scraping tools (e.g., Beautiful Soup, Scrapy) and APIs. The collected data is then normalized and its format is checked for quality. This step also involves removing duplicate data and imputing missing values. The input data is obtained from public data sources, and the output data is normalized case data. The data is then stored in a database (e.g., MySQL, PostgreSQL, MongoDB). An index is built for the stored data to enable efficient access for later search and analysis.

[0759] Step 2: Enter case information and recognize emotional state

[0760] Users use a device (e.g., PC, tablet, or smartphone) to enter information related to their legal case. Specifically, they fill out an input form with the details of the lawsuit, the desired outcome, and supporting documents. At the same time, the emotion engine uses the device's camera and microphone to analyze the user's facial expressions and voice to recognize their emotional state in real time. Emotional states include stress levels, anxiety, and joy. The input data are the lawsuit information and emotion data provided by the user, and the output data are the analyzed lawsuit information and emotion state.

[0761] Step 3: Submit your case information

[0762] The terminal transmits the entered legal information and the emotion data recognized by the emotion engine to the server. This transmission is performed using encrypted communication (e.g., HTTPS). The input data is the legal information and emotion data, and the output data is the data securely transmitted to the server.

[0763] Step 4: Find similar cases

[0764] The server searches for similar past cases in the database based on the received user's lawsuit information and emotion data. To do this, it uses natural language processing (NLP) algorithms (e.g., BERT, Word2Vec) and similarity calculation algorithms (e.g., Cosine Similarity). The input data are the received lawsuit information and emotion data, and the output data are the searched similar cases.

[0765] Step 5: Calculating Win Rate

[0766] The server uses data extracted from similar cases to calculate the probability of winning a lawsuit using a machine learning model (e.g., logistic regression, random forest). This step also takes into account the user's emotional data. For example, if the user is very nervous, the predictive model will compensate for that effect. The input data are similar cases and emotional data, and the output data is the calculated probability of winning a lawsuit.

[0767] Step 6: Generate reports

[0768] The server automatically generates a report containing the winning probability calculation results and details of related similar cases. This report includes detailed explanations to maintain transparency of the calculation process and also serves the purpose of reducing the mental burden on the user. The input data are the winning probability calculation results and detailed data of similar cases, and the output data is the generated report.

[0769] Step 7: Provide a report

[0770] The terminal displays the generated report to the user. The user interface is designed to be intuitive. The user views the report through the terminal and decides whether to proceed with the lawsuit. Feedback from the emotion engine is also provided here. The input data is the generated report, and the output data is the report presentation and feedback to the user.

[0771] Step 8: Litigation training of generative AI

[0772] When a legal amendment is made, the server collects data on court records and lawyers' legal activities and trains this data into a generative AI model (e.g., GPT-3, BERT). The input data is the revised court records and legal activities data, and the output data is the model data trained by the generative AI.

[0773] Step 9: Defending the generative AI

[0774] When a user requests a defense from the generative AI, they re-enter the necessary legal information into their device and send the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the generative AI. The server then formulates an optimal defense strategy based on the received information and learning data. The input data is the re-entered legal information and emotion data, and the output data is the defense strategy created by the generative AI.

[0775] Step 10: Real-time advocacy

[0776] Once the lawsuit begins, the generative AI will actually conduct the defense in court. It will provide real-time advice to the user and support the trial, including suggesting when to present evidence and what questions to ask. It will continuously monitor the user's emotional state using an emotion engine, and the generative AI will adjust its strategy as needed. The input data is the user's real-time emotional state and the status of the trial, and the output data is real-time advice and support from the generative AI.

[0777] (Application example 2)

[0778] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0779] While conventional litigation support systems can calculate the probability of winning based on past case data, they do not take into account the user's emotional state and lack support for mental stress and anxiety. Furthermore, because they do not provide user support or emotional monitoring in real time during legal proceedings, they do not alleviate the psychological burden on users in the courtroom. This hinders user decision-making and effective legal proceedings during legal proceedings.

[0780] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0781] In this invention, the server includes means for collecting and storing case data from past civil and family law cases in a database; means for receiving case information entered by a user and searching the database for similar past cases; means for calculating the success rate of the case based on the searched similar cases and providing the result to the user; means for recognizing the user's emotional state using an emotion engine and more accurately calculating the success rate of the case taking the emotional state into account; and means for generating and providing a report that takes the user's emotions into account. This enables accurate calculation of the success rate of the case taking the user's emotional state into account and also provides psychological support for making litigation decisions. Furthermore, by monitoring the user's emotions and providing mental support during real-time legal proceedings, the server can reduce the user's psychological burden and support the effective progress of the case.

[0782] A "civil lawsuit" is a lawsuit to resolve legal disputes between private citizens, where individuals or entities dispute each other's rights and obligations.

[0783] A "family litigation" is a judicial proceeding to resolve disputes about domestic issues and relationships.

[0784] "Case law data" is a record of past decisions made by courts on legal issues.

[0785] A "database" is a system for efficiently searching, storing, and managing large amounts of organized data.

[0786] "Similar cases" are past court decisions that have similar facts and legal issues to those in the current lawsuit.

[0787] A "success rate" is the percentage of chances of success in a lawsuit.

[0788] "Generative AI" is artificial intelligence that generates and analyzes new information based on machine learning and deep learning.

[0789] An "emotion engine" is a system that recognizes and analyzes human emotions from voice, facial expressions, context, etc.

[0790] A "report" is a report that summarizes analysis results and information.

[0791] A "trial record" is a record of all documents and evidence arising in the course of a trial.

[0792] "Advocacy" refers to the work and actions of a lawyer to assist and defend a client under the law.

[0793] A "user interface" is a screen or operating means that allows a user to interact with a system or application.

[0794] "Mental support" refers to services that help users to alleviate and support their mental and emotional anxieties and difficulties.

[0795] This invention is a system that calculates the success rate of lawsuits based on past legal precedent data, and combines advocacy activities using generative AI with user support using an emotion engine. This system consists of a server, terminals, users, and an emotion engine.

[0796] Data collection and storage

[0797] The server collects historical civil and family law case data from legal-related public data sources and open databases using web scraping and APIs, normalizes the data, and stores it in a database that is managed for efficient searching and referencing.

[0798] Entering lawsuit information and calculating win rate

[0799] Using the terminal, the user inputs information related to their legal case. During this information input process, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The emotional state and legal case information are then sent to the server.

[0800] The server searches a database based on the received lawsuit information and sentiment data to extract similar past lawsuits. This search uses natural language processing (NLP) algorithms and similarity calculations. Using the analysis results of the extracted similar cases and sentiment data, a machine learning model (e.g., logistic regression or random forest) is used to calculate the winning rate.

[0801] Calculating winning rates and providing results

[0802] The server uses the emotion engine to take into account the user's emotional state and more accurately calculate the success rate of the lawsuit. A report is generated containing the calculation results and details of related similar cases. The report is displayed to the user through the user interface on their device, allowing them to view the success rate, similar cases, and specific advice. Based on the report and advice provided, the user decides whether to proceed with the lawsuit. Feedback from the emotion engine can also be used to help make legal decisions.

[0803] Generative AI advocacy

[0804] When legal amendments are made or new case data is added, the server collects court records and data on lawyers' legal activities and trains the generative AI. Based on the learned data, the generative AI acquires legal skills to support litigation.

[0805] When a user requests the Generative AI to defend their case, they re-enter the necessary legal information into their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects that information in the Generative AI. The server then formulates an optimal defense strategy based on the received information and learning data. Once the lawsuit begins, the Generative AI will conduct defense activities in court and provide advice to the user in real time. The emotion engine continuously monitors the user's emotional state, and the Generative AI will adjust its strategy as needed.

[0806] Specific examples

[0807] For example, if a company reports a data breach incident, they can use the AI ​​security assistant application to input details of the incident. The emotion engine analyzes the stress and anxiety of the person in charge and compares it with similar cases in the database to conduct a risk assessment. The results are provided as a report, and any necessary mental support is also provided.

[0808] Example prompt sentence:

[0809] "Important data was leaked by an internal employee. As a countermeasure, we are considering strengthening log management and providing employee training."

[0810] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0811] Step 1:

[0812] The server collects past civil and family law case data from legal public data sources and open databases, using web scraping and APIs, and stores it in a database. The data is then normalized and stored in the database.

[0813] Input: Public legal data sources

[0814] Output: Normalized case data

[0815] Step 2:

[0816] The user uses the device to input information related to their legal case (e.g., the details of the case, the desired outcome, and supporting documents). As the user inputs information, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state.

[0817] Input: Lawsuit case information, user facial expression and voice data

[0818] Output: Input case information, perceived emotional state

[0819] Step 3:

[0820] The server searches the database based on the received user's lawsuit information and emotion data to extract similar past lawsuits, using natural language processing (NLP) algorithms and similarity calculations.

[0821] Input: User's lawsuit information, emotional data

[0822] Output: Extracted similar lawsuit cases

[0823] Step 4:

[0824] Based on the extracted similar lawsuits and sentiment data, the server uses machine learning models to calculate the probability of winning a lawsuit, using algorithms such as logistic regression and random forests.

[0825] Input: Similar lawsuit cases, emotional data

[0826] Output: Calculated success rate of lawsuits

[0827] Step 5:

[0828] The server further adjusts the chances of winning the case by taking into account the emotional state of the user as recognized using the emotion engine.

[0829] Inputs: Calculated case win rate, perceived emotional state

[0830] Output: Adjusted lawsuit win rate

[0831] Step 6:

[0832] The server generates a report containing the results of the calculation and details of related similar cases, which can be transmitted to the terminal and viewed by the user through the user interface.

[0833] Input: Adjusted case win rate, details of related similar cases

[0834] Output: Generated report

[0835] Step 7:

[0836] The user decides whether to proceed with the lawsuit based on the report and advice provided through the terminal. During this decision-making process, the emotion engine monitors the user's emotional state and provides feedback.

[0837] Input: Report, Advice

[0838] Output: User's legal decision

[0839] Step 8:

[0840] When legal amendments or new case law data are added, the server collects court records and data on lawyers' legal activities and trains the generation AI.

[0841] Input: Latest court records and advocacy data

[0842] Output: A trained generative AI model

[0843] Step 9:

[0844] When a user requests the AI ​​to represent them, they re-enter the case information into their device, which is then sent to the server. During this process, the emotion engine recognizes the user's emotions and reflects them in the AI.

[0845] Input: Re-entered case information, perceived emotional state

[0846] Output: Data passed to the generative AI

[0847] Step 10:

[0848] The server uses the trained generative AI model to develop an optimal defense strategy and provides real-time advice to the user during the courtroom. The emotion engine continuously monitors the user's emotional state, and the generative AI adjusts the strategy.

[0849] Input: Generative AI model, real-time user emotion data

[0850] Output: Real-time advice and tailored defense strategies

[0851] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0852] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0853] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0854] [Third embodiment]

[0855] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0856] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0857] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0858] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0859] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0860] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0861] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0862] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0863] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0864] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0865] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0866] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0867] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data of civil and family law lawsuits, and furthermore, uses generative AI to conduct defense activities in court. This system operates in cooperation with three parties: a server, a terminal, and a user.

[0868] Data collection and storage

[0869] The server periodically collects case data for past civil and family law cases from legal-related public data sources and open databases. This data includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[0870] Entering lawsuit information and calculating win rate

[0871] The user uses the terminal to input information related to their legal case, including the specifics of the case, the desired outcome, and supporting documents. This input data is then sent from the terminal to the server.

[0872] The server searches the database based on the received user's lawsuit information to extract similar past cases. This is done using natural language processing (NLP) algorithms and similarity calculations. Based on the analysis results of the extracted similar cases, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the lawsuit. The calculation results are generated in report format and sent to the terminal.

[0873] The user can review the reports provided by the terminal and view details of the winning rate and related cases. Based on this information, the user can decide whether to proceed with the lawsuit.

[0874] Generative AI advocacy

[0875] When legal changes are made, the server collects data on court records and lawyers' legal activities and trains the generative AI. This training data includes minutes, evidence, and legal content. The generative AI acquires legal skills to support litigation based on the learned data.

[0876] When a user requests that the generated AI represent them, they input the necessary information into their device and send a request for the generated AI to the server. The server then uses this information and learning data to formulate an optimal defense strategy. When the generated AI represents its case in court, it provides real-time advice to the user and supports the progress of the trial.

[0877] Users can monitor the AI's defense activities in court and provide instructions as necessary. Once the outcome of the lawsuit is known, a detailed report will be provided via the device, along with advice on future actions.

[0878] Specific examples

[0879] For example, suppose a user is considering divorce proceedings and is fighting over asset division and child custody. The user uses a terminal to enter the details of the lawsuit and supporting documents. The server receives this information and searches its database for similar past divorce cases. Based on the search results, the server calculates a 70% chance of success and provides a report to the user.

[0880] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend their case. The generated AI will study past court records and conduct a defense in court. Finally, if the generated AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0881] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment in which ordinary households can easily receive legal support.

[0882] The processing flow will be explained below.

[0883] Step 1:

[0884] The server collects historical civil and family law case data from legal public data sources and open databases, using web scraping and APIs to normalize and retrieve the data.

[0885] Step 2:

[0886] The server preprocesses the collected case data, converting it into a standard format before storing it in a database, including deduplication, filling in missing values, and text cleansing.

[0887] Step 3:

[0888] The user uses a terminal to enter information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents). The input form has required and optional fields, and the user must enter all required information.

[0889] Step 4:

[0890] The terminal transmits the entered litigation information to the server, where the transmitted data is encoded and sent using a secure communication method.

[0891] Step 5:

[0892] The server analyzes the received user's lawsuit information and searches for similar past cases in its database, using natural language processing (NLP) algorithms and similarity calculations.

[0893] Step 6:

[0894] The server calculates the probability of winning a lawsuit based on the data of similar cases retrieved, using machine learning models (such as logistic regression and random forests) to take into account necessary variables.

[0895] Step 7:

[0896] The server generates a report detailing the winning probability calculation and related similar cases, including a detailed explanation of the calculation process to ensure transparency.

[0897] Step 8:

[0898] The server sends the generated reports to the terminal, which are presented in a visually easy-to-understand format (graphs, tables, text).

[0899] Step 9:

[0900] The terminal displays the submitted report to the user through a user interface, where the user can view the win rate, similar cases, and specific advice.

[0901] Step 10:

[0902] Users can decide whether to proceed with a lawsuit based on the reports and advice provided, and can also ask additional questions or complete the data.

[0903] Step 11:

[0904] When legal changes are made, the server collects data on court records and lawyers' legal activities, including minutes, evidence, and content of legal arguments, and organizes it into a learning dataset.

[0905] Step 12:

[0906] The server trains the generative AI using data based on court records, using deep learning techniques to train the AI ​​to acquire effective advocacy skills.

[0907] Step 13:

[0908] To request the generated AI to defend the case, the user re-enters the necessary legal information into the device and submits the request.

[0909] Step 14:

[0910] The server receives a request for legal representation from the user and formulates an optimal legal representation strategy based on the learning data, and if necessary, communicates with the user.

[0911] Step 15:

[0912] Once litigation begins, the generative AI will actually act as a defense in court, providing real-time advice to the user and supporting the proceedings, including suggesting when to present evidence and what questions to ask.

[0913] Step 16:

[0914] After the verdict is reached, the server analyzes the results and generates a detailed report that is sent to the device, including the outcome of the verdict and advice on future actions.

[0915] Step 17:

[0916] The user will use the provided report to determine any necessary follow-up actions (eg, appeal decisions, etc.).

[0917] The above are the specific processing steps of the entire system.

[0918] Example 1

[0919] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0920] Existing legal support systems have difficulty calculating an appropriate win rate for each user's legal information, or providing real-time support for legal activities based on past cases. Furthermore, to efficiently support legal activities in court, it is necessary to analyze a huge amount of past court records and lawyers' legal activity data, and use generative AI to formulate optimal legal strategies.

[0921] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0922] In this invention, the server includes means for collecting past litigation case data from legal-related public data sources and open databases and storing it in a database, means for receiving litigation information entered by a user and searching the database for similar past cases using a natural language processing algorithm, means for analyzing the searched similar cases and calculating the success rate of the litigation using a machine learning model and providing the results to the user in the form of a report, and means for the user to use a terminal for inputting litigation information and evidential materials. This allows users to calculate the success rate of their own litigation cases with high accuracy and to develop optimal defense strategies based on similar past cases.

[0923] "Legal-related public data sources" refers to databases of laws, precedents, court records, etc. that are publicly available from governments and related agencies.

[0924] "Open database" refers to a database provided by a third party that is freely accessible on the Internet.

[0925] "Litigation case data" refers to a collection of data including records of past civil and family law litigation, judgments, related laws and regulations, evidence lists, and background information on the parties involved.

[0926] "Database" refers to an information system for efficiently storing, retrieving, and managing information.

[0927] "Law case information entered by the user" refers to the specific details, desired outcome, supporting documents, etc., of the user's own legal case.

[0928] "Natural language processing algorithm" refers to technology that allows computers to analyze human language and understand its meaning.

[0929] "Similar past cases" refers to past legal cases that have similar characteristics based on specific criteria to the user's legal information.

[0930] A "machine learning model" refers to a statistical algorithm that uses data to learn and make predictions, classifications, and other similar tasks.

[0931] "Generative AI" refers to artificial intelligence that can learn from past data and automatically perform specific tasks.

[0932] A "defense strategy" refers to a plan or method that aims to gain an advantage in a lawsuit based on law and evidence.

[0933] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data for civil and family law lawsuits, and uses generative AI to conduct defense activities in court. This system operates in cooperation with three parties: a server, a terminal, and a user.

[0934] Data collection and storage

[0935] The server periodically collects past civil and family law case data from legal-related public data sources and open databases using curl or API. The collected data is then stored in a database (e.g., MySQL or PostgreSQL) after checking its accuracy. It is also indexed to streamline data search and management.

[0936] Entering lawsuit information and calculating win rate

[0937] The user enters information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents) into a dedicated form on their device. The device temporarily stores the entered data, and once all the data is collected, it sends an HTTP POST request to the server.

[0938] The server uses a natural language processing (NLP) algorithm to search its database for similar past cases based on the received user's lawsuit information. Specifically, it generates a search query and performs a full-text search of the past lawsuit database. Based on the analysis results of the extracted similar cases, it uses a machine learning model (e.g., logistic regression or random forest) to calculate the success rate of the lawsuit.

[0939] The results are formatted into a report and sent from the server to the terminal. The user can then review the report provided by the terminal and refer to the winning rate and details of related cases. Based on this information, the user can decide whether to proceed with the lawsuit.

[0940] Generative AI advocacy

[0941] When legal changes or new legal precedents arise, the server collects court records and lawyers' advocacy data and trains the generative AI using a natural language processing framework (e.g., BERT). This training data includes court records, evidence, and advocacy content.

[0942] When a user requests a defense from the generated AI, they input the necessary information into their device and send a defense request to the server. The server receives this information and provides the necessary data to the generated AI.

[0943] The generative AI will develop an optimal defense strategy based on past learning data and support legal actions in court. It will send advice to the user in real time to support the trial. The user can monitor the generative AI's legal actions in court and give instructions as necessary. Once the outcome of the trial is known, a detailed report will be provided to the user via their device, along with advice on future countermeasures.

[0944] Specific examples

[0945] For example, suppose a user is considering divorce proceedings and is fighting over asset division and child custody. The user uses a terminal to enter the details of the lawsuit and supporting documents. The server receives this information and searches its database for similar past divorce cases. Based on the search results, the server calculates a 70% chance of success and provides a report to the user.

[0946] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend their case. The generated AI will learn from past court records and conduct a defense in court. Finally, if the generated AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[0947] An example prompt is, "I am currently in a divorce proceeding regarding asset division and child custody. How can I calculate my chances of winning based on past case data and use generative AI to defend my case in court?"

[0948] The above is a specific embodiment of the present invention. This system will reduce the cost and time required for litigation, and create an environment in which ordinary households can easily receive legal support.

[0949] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0950] Step 1:

[0951] The server periodically collects past civil and family law case data from legal-related public data sources and open databases via the Internet, using curl and APIs to retrieve the data.

[0952] Input: Data source URL or API endpoint

[0953] Data processing: Formatting the acquired data and checking its accuracy

[0954] Output: A formatted dataset

[0955] Specific example: "curl https: / / example-legal-database.com / api / cases"

[0956] Step 2:

[0957] The server stores the collected data in a database (e.g., MySQL or PostgreSQL) and indexes it for efficient searching.

[0958] Input: Formatted dataset

[0959] Data manipulation: Inserting and indexing data based on the database schema

[0960] Output: Data stored in the database

[0961] Specific example: "INSERT INTO cases_table (case_id, case_type, verdict, laws, evidence, parties_info) VALUES (...)"

[0962] Step 3:

[0963] The user enters the case information into a dedicated form on the terminal.

[0964] Input: Contents of lawsuit, desired outcome, evidence, etc.

[0965] Data calculation: Temporarily save input data

[0966] Output: Temporarily saved dataset

[0967] Specific example: "Enter the details of the lawsuit in text on a web form"

[0968] Step 4:

[0969] The terminal sends the completed input data to the server via an HTTP POST request.

[0970] Input: Completed dataset

[0971] Data processing: Generating the request body

[0972] Output: Request sent to server

[0973] Example: "POST / api / case_submission { case_data: {...}}"

[0974] Step 5:

[0975] Based on the user's lawsuit information received by the server, a natural language processing (NLP) algorithm is used to search for similar past cases.

[0976] Input: User's case information

[0977] Data Computing: Database Querying Using Full-Text Search Algorithms

[0978] Output: List of similar jobs

[0979] Specific example: SELECT FROM cases_table WHERE MATCH (case_text) AGAINST ('User's lawsuit details')

[0980] Step 6:

[0981] The server analyzes data from similar cases and uses machine learning models (such as logistic regression or random forest) to calculate the probability of winning a lawsuit.

[0982] Input: Data for similar cases

[0983] Data Computation: Probability Computation Using Machine Learning Models

[0984] Output: Win rate calculation result

[0985] Specific example: "Calculating probability using a machine learning model (e.g., sklearn's LogisticRegression.predict_proba)"

[0986] Step 7:

[0987] The server formats the winning rate calculation results into a report format and sends it to the terminal.

[0988] Input: Win rate calculation result

[0989] Data processing: Report generation (e.g. HTML or PDF format)

[0990] Output: Report sent to user terminal

[0991] Specific example: "Using a report generation library (e.g. ReportLab)"

[0992] Step 8:

[0993] The user reviews the report on the terminal and decides whether to proceed with the lawsuit.

[0994] Input: Report data

[0995] Data calculation: Understanding and evaluating report content

[0996] Output: Decision on next action

[0997] Specific example: "Display report PDF on browser"

[0998] Step 9:

[0999] The user inputs the necessary information to request legal representation from the generated AI and sends it from the terminal to the server.

[1000] Input: Detailed information about the legal request

[1001] Data processing: Generating request data

[1002] Output: Request sent to server

[1003] Specific example: "Enter legal request information on a web form"

[1004] Step 10:

[1005] The server provides the received request information to the generation AI, which then formulates the optimal defense strategy.

[1006] Input: Legal request information

[1007] Data Computation: Strategy Planning with Generative AI

[1008] Output: Strategic proposal

[1009] Specific example: "API call to pass request information to the generation AI"

[1010] Step 11:

[1011] Generative AI will act as a defense in court and provide real-time advice to users.

[1012] Input: Strategic data generated by generative AI

[1013] Data Computing: Real-time Advice Generation

[1014] Output: Provide advice to the user

[1015] Specific example of operation: "Generative AI sends advice in real time in chat format"

[1016] Step 12:

[1017] The user will monitor the generated AI's defense activities in court and provide instructions as necessary.

[1018] Input: Generative AI advice and on-site situation

[1019] Data Computing: Advice and Situation Assessment

[1020] Output: Sending instructions

[1021] Specific example: "Real-time feedback and instructions on web applications"

[1022] Step 13:

[1023] When the outcome of the lawsuit becomes clear, the server generates a detailed report and sends it to the device, along with advice on future countermeasures.

[1024] Input: Lawsuit outcome data

[1025] Data processing: detailed report generation and analysis of future countermeasures

[1026] Output: Detailed report and further advice to the user

[1027] Specific example: "Generating and sending a final result report"

[1028] (Application example 1)

[1029] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1030] In the past, in civil and family law cases, it was difficult to calculate the success rate of a case based on past case data, and there was a lack of means to provide real-time advice to properly support legal advocacy in court. As a result, in many cases, users found it difficult to make informed decisions, and the quality of legal advocacy was inconsistent.

[1031] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1032] In this invention, the server includes means for collecting case data on past civil and family law litigation cases and storing it in a database, means for receiving litigation information entered by a user and searching the database for similar past cases, means for calculating the success rate of the litigation based on the searched similar cases and providing the result to the user in real time, and means for receiving the litigation information entered by a user via a smartphone and calculating the success rate of the litigation in real time. This allows the user to obtain a highly accurate litigation success rate based on past cases and to obtain the necessary information in real time via their smartphone.

[1033] "Past civil and family law case precedent data" refers to past civil and family law case precedent information collected from public institutions and legal databases.

[1034] "Means of storing in a database" refers to a method or system for systematically organizing collected case law data and storing it so that it can be easily searched and analyzed.

[1035] "Means for receiving litigation information entered by the user and searching for similar past cases" refers to technologies and algorithms for obtaining information about litigation provided by the user and using that information to find similar cases from past case data.

[1036] "Means for calculating the success rate of a lawsuit and providing the result to the user in real time" refers to technology or methods that analyze past case law data and information provided by the user to calculate the likelihood that the outcome of the lawsuit will be favorable to the user and present the result to the user in real time.

[1037] "Means of obtaining necessary information in real time via a smartphone" refers to applications and systems that allow users to receive information provided in real time via a smartphone, which is a mobile device.

[1038] "Means for training generative AI" refers to methods and techniques for collecting data on court records and legal actions and using them to train generative AI models.

[1039] "Means for inputting information to request defense from the generated AI" refers to an interface through which a user can input the information necessary to request defense from the generated AI.

[1040] "Means for generative AI to conduct legal advocacy in court" refers to technologies and systems that enable generative AI to conduct legal advocacy in court based on data learned by the AI ​​and provide legal advice in real time.

[1041] "Means for visually displaying through a user interface" refers to an interface that allows a user to visually check the results of the lawsuit's winning probability calculation and details of similar precedents.

[1042] "Means for providing advice to help users decide whether to proceed with litigation" refers to a system or method for providing information or advice to users that will help them decide whether to proceed with litigation.

[1043] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data for civil and family law cases, and then uses generative AI to conduct defense activities in court. This system operates in cooperation with a server, a terminal, and a user.

[1044] Data collection and storage

[1045] First, the server periodically collects case data on past civil and family law cases from public institutions and legal databases. This data includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties involved. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[1046] Entering lawsuit information and calculating win rate

[1047] Users use a device (such as a smartphone) to enter information related to their legal case. This information includes the specific details of the case, the desired outcome, and supporting documents. The entered data is sent from the device to a server. The server then searches a database based on the user's legal case information and extracts similar past cases. This process uses natural language processing (NLP) algorithms and similarity calculations. Based on the analysis of the extracted similar cases, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the case. The calculation results are generated in the form of a report and sent to the device.

[1048] Generative AI advocacy

[1049] When legal amendments are made, the server collects data on court records and lawyers' legal activities and trains the generated AI. This training data includes minutes, evidentiary documents, and legal content. The generated AI uses the trained data to acquire legal skills to support litigation. When a user requests the generated AI to act as a legal representative, they enter the necessary information on their device and send a request for legal representation to the server. The server then formulates the optimal legal strategy based on this information and the training data. When the generated AI acts as a legal representative in court, it sends advice to the user in real time to support the progress of the trial. The user can check the generated AI's legal activities in court and issue instructions as necessary. Once the outcome of the lawsuit is known, a detailed report is provided via the device, along with advice on future countermeasures.

[1050] User Interface

[1051] The calculation results of the winning rate and details of similar cases sent from the server are visually displayed through the user interface, making it easier for users to intuitively understand the lawsuit information. The same interface also provides advice to help users decide whether to proceed with the lawsuit.

[1052] Specific examples

[1053] For example, suppose a user is considering divorce proceedings and is fighting over asset distribution and child custody. The user uses a terminal to input the details of the lawsuit and supporting documents. The server receives this information and searches a database for similar past divorce cases. Based on the search results, it calculates a 70% chance of success and provides a report to the user. If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend the case. The generated AI has learned from past court records and will defend the case in court. Finally, if the generated AI's defense results in a favorable verdict, the user reviews the results and decides what subsequent actions are required.

[1054] Examples of prompts for generative AI models

[1055] If you are considering divorce proceedings and are disputing asset distribution and child custody, please analyze past legal precedents related to this type of case and propose legal action that focuses on the fairness of asset distribution and the best interests of the children.

[1056] The above is a specific embodiment of the present invention.

[1057] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1058] Step 1: Collect and store data

[1059] The server periodically collects case data for past civil and family law cases from legal-related public data sources and open databases. The data input includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties. This data is stored in a database and managed for easy searching and reference.

[1060] Step 2: Enter your case information

[1061] Users use their devices (smartphones) to input specific information about their legal case—the nature of the case, the desired outcome, evidence, etc.—and this input data is sent from the device to the server.

[1062] Step 3: Find similar cases

[1063] The server searches the database based on the user's lawsuit information received in step 2 to extract similar past cases. This uses natural language processing (NLP) algorithms and similarity calculations. Specifically, the entered lawsuit content is tokenized, vectorized using techniques such as TF-IDF, and compared with similar cases in the database.

[1064] Step 4: Calculating the probability of success of your lawsuit

[1065] The server uses a machine learning model (such as logistic regression or random forest) to calculate the success rate of the lawsuit based on the analysis results of similar cases extracted in step 3. To do this, the input data and statistical data of similar cases are plotted and input into the model to predict the success rate. The calculation results are generated in report format and sent to the terminal.

[1066] Step 5: Litigation Support Report

[1067] The terminal receives the report sent from the server in step 4. The report includes details of similar cases along with the calculated win rate. The user can view this information and use it to decide whether to proceed with the lawsuit.

[1068] Step 6: Generative AI learns to advocate

[1069] When legal changes occur, the server collects court records and data on lawyers' legal activities and trains the AI. This training data includes minutes, evidence, and legal content. The AI ​​uses the collected data to acquire legal skills.

[1070] Step 7: Enter your request for legal representation

[1071] When a user requests a generated AI to defend them, they input the necessary information into their device and send the request to the server, including the specific details of the lawsuit and the key points of the defense they are seeking.

[1072] Step 8: Developing a defense strategy

[1073] The server formulates an optimal defense strategy based on the user information received in step 7 and the data learned by the generating AI. The generating AI utilizes the learned data to provide real-time advice to the user for advocacy in court.

[1074] Step 9: Real-time legal support

[1075] As users pursue their cases in court, the generative AI provides real-time support for their defense, providing advice on how to present evidence and what to say. Users can use this advice to make decisions in court.

[1076] Step 10: Reporting the outcome of the case

[1077] Once the case is over, the device receives a detailed report, including the outcome of the case and the effectiveness of the advice provided by the AI. The user can use this information to decide on future actions.

[1078] These are the specific processing steps of the system. Based on the specific software and algorithms used in each step, we can clearly understand the entire process from data input to output.

[1079] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1080] This invention combines a system that calculates the success rate of lawsuits based on past precedent data from civil and family law lawsuits, and conducts legal defense activities in court using generative AI, with an emotion engine that recognizes the user's emotions. This system consists of a server, a terminal, a user, and an emotion engine.

[1081] Data collection and storage

[1082] The server periodically collects past civil and family law case data from legal-related public data sources and open databases. This data is obtained by normalizing it using web scraping and APIs. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[1083] Entering lawsuit information and calculating win rate

[1084] The user uses the terminal to input information related to their legal case (e.g., the lawsuit content, desired outcome, and supporting documents). During the process of inputting this information, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The emotion recognition results are sent to the server along with the input information.

[1085] The server searches the database based on the received user's lawsuit information and sentiment data to extract similar past cases. This is done using natural language processing (NLP) algorithms and similarity calculations. Based on the analysis results of the extracted similar cases and sentiment data, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the lawsuit.

[1086] Calculating winning rates and providing results

[1087] The server uses an emotion engine to take into account the user's recognized emotional state to more accurately calculate the probability of winning the lawsuit. A report is generated containing the calculation results and details of related similar cases. The report includes detailed explanations to ensure transparency of the calculation process and is presented to the user in a manner that minimizes mental burden.

[1088] The terminal displays the submitted report to the user through a user interface. The user can view the win rate, similar cases, and specific advice. Based on the provided report and advice, the user decides whether to proceed with the lawsuit. Feedback from the emotion engine can also be used to help make a lawsuit decision.

[1089] Generative AI advocacy

[1090] When legal changes are made, the server collects data on court records and lawyers' legal activities and trains the generative AI on this data. This data includes minutes, evidence, and legal content, and is compiled as a training dataset. The generative AI uses the learned data to acquire legal skills to support litigation.

[1091] When a user requests the AI ​​to represent them, they re-enter the necessary legal information into their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the AI. The server then formulates the optimal defense strategy based on the received information and learning data.

[1092] Once litigation begins, the generative AI will actually conduct the defense in court, providing real-time advice to the user and supporting the trial, including suggesting when to present evidence and what questions to ask. Using an emotion engine, the generative AI will continuously monitor the user's emotional state and adjust its strategy as needed.

[1093] Specific examples

[1094] For example, suppose a user is considering divorce proceedings and is battling over asset distribution and child custody. The user uses a device to input the details of the lawsuit and supporting documents. At this time, the emotion engine recognizes emotions such as stress and anxiety from the user's facial expressions and voice. The server receives this information and searches its database for similar past divorce lawsuit cases. Based on the search results, taking emotions into account, the server calculates a 70% chance of success and provides a report to the user.

[1095] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the Generator AI defend them. The emotion engine again recognizes the user's emotions and reflects them in the Generator AI. The Generator AI then conducts the defense in court and provides real-time advice to the user. Finally, if the Generator AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[1096] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment where ordinary households can easily receive legal support. Furthermore, by providing support that takes the user's emotions into consideration, it is possible to make more appropriate legal decisions while reducing the mental burden.

[1097] The processing flow will be explained below.

[1098] Step 1:

[1099] The server collects historical civil and family law case data from legal public data sources and open databases, using web scraping and APIs to normalize and retrieve the data.

[1100] Step 2:

[1101] The server preprocesses the collected case data, converting it into a standard format before storing it in a database, including deduplication, filling in missing values, and text cleansing.

[1102] Step 3:

[1103] Users use a terminal to input information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents). During this input process, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state.

[1104] Step 4:

[1105] The terminal transmits the entered lawsuit information and the user's emotion data recognized by the emotion engine to the server. The transmitted data is encoded and sent using a secure communication method.

[1106] Step 5:

[1107] The server analyzes the received user's lawsuit information and sentiment data and searches for similar past cases in its database, using natural language processing (NLP) algorithms and similarity calculations.

[1108] Step 6:

[1109] The server calculates the probability of winning a lawsuit based on the data of similar cases it has searched for, using machine learning models (such as logistic regression or random forest) and taking into account necessary variables.

[1110] Step 7:

[1111] The server adjusts the probability of winning a lawsuit by taking into account the user's emotional state, which is recognized using an emotion engine. For example, if the user is in a high stress state, it is taken into account as a risk factor.

[1112] Step 8:

[1113] The server generates a report detailing the winning probability calculation and related similar cases, including detailed explanations to ensure transparency of the calculation process.

[1114] Step 9:

[1115] The server sends the generated reports to the terminal, which are presented in a visually easy-to-understand format (graphs, tables, text).

[1116] Step 10:

[1117] The terminal displays the submitted report to the user through a user interface, where the user can view the win rate, similar cases, and specific advice.

[1118] Step 11:

[1119] Users can decide whether to proceed with a lawsuit based on the reports and advice provided, and feedback from an emotion engine can also be used to help with the lawsuit decision.

[1120] Step 12:

[1121] When legal changes are made, the server collects data on past court records and lawyers' legal activities, including minutes, evidence, and content of arguments, and uses this data as training data for the generative AI.

[1122] Step 13:

[1123] The server trains the generative AI using data based on court records, using deep learning techniques to train the AI ​​to acquire effective advocacy skills.

[1124] Step 14:

[1125] To request the AI ​​to defend the case, the user re-enters the necessary legal information into the device and submits the request. At this time, the emotion engine again recognizes the user's emotions and reflects them in the AI.

[1126] Step 15:

[1127] The server receives a request for legal representation from the user and formulates an optimal legal representation strategy based on the learning data, and if necessary, communicates with the user.

[1128] Step 16:

[1129] Once litigation begins, the generative AI will actually act as a defense in court, providing real-time advice to the user and supporting the proceedings, including suggesting when to present evidence and what questions to ask.

[1130] Step 17:

[1131] During courtroom advocacy, the generative AI uses an emotion engine to continuously monitor the user's emotional state and adjust its strategy as needed.

[1132] Step 18:

[1133] After the verdict is reached, the server analyzes the results and generates a detailed report that is sent to the device, including the outcome of the verdict and advice on future actions.

[1134] Step 19:

[1135] The user will use the provided report to determine any necessary follow-up actions (eg, appeal decisions, etc.).

[1136] The above are the specific processing steps of the entire system.

[1137] Example 2

[1138] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1139] Conventional litigation support systems could calculate the probability of winning a lawsuit based on past case data, but they were unable to take the user's emotional state into account, resulting in low accuracy in the calculation of the win rate. Furthermore, they were limited in their use of court records and data on lawyers' legal activities, resulting in insufficient courtroom advocacy by the generated AI. Furthermore, they required time and effort to input legal information, and lacked a means to visually present detailed case information. This made it difficult for users to receive appropriate legal support and placed a significant mental burden on them.

[1140] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1141] In this invention, the server includes means for collecting case data on past civil and family law lawsuits and storing it in a database, means for receiving lawsuit information entered by a user and searching the database for similar past cases, means for calculating the success rate of the lawsuit based on the searched similar cases and providing the result to the user, means for analyzing the user's facial expressions and voice and recognizing their emotional state, and means for calculating the success rate by reflecting the emotional state in the calculation means. This enables highly accurate calculation of the success rate that takes the user's emotional state into account, enabling appropriate defense activities in court using generative AI, and providing comprehensive legal support and reducing the mental burden for the user.

[1142] "Past civil and family law case data" refers to a collection of past court decisions and related documents relating to civil and family law cases.

[1143] A "database" is a data management system that systematically stores collected case law data and related information and allows for searching and inquiry.

[1144] "User" refers to an individual or company that utilizes the system to enter information about a lawsuit and receive assistance.

[1145] "Lawsuit information" is information related to the lawsuit entered by the user, including the details of the lawsuit, the desired outcome, and supporting documents.

[1146] "Similar past cases" refers to past precedents that have elements or outcomes similar to the current litigation case.

[1147] "Win rate" is the percentage of the likelihood of success in a lawsuit, calculated based on past similar cases and current litigation information.

[1148] "Analyzing facial expressions and voice to recognize emotional states" refers to the process of reading the user's emotions at that time by analyzing their facial expressions and tone of voice.

[1149] "Generative AI" refers to an artificial intelligence model that learns from past data and automatically responds and makes suggestions to new situations.

[1150] "Courtroom advocacy" refers to a series of activities carried out by lawyers and generative AI in court, such as advocacy, advice, presenting evidence, and proposing questions.

[1151] A "user interface" is a screen or mechanism that allows a user to operate a system, input information, and view results.

[1152] This invention combines a system that calculates the success rate of lawsuits based on past precedent data from civil and family law lawsuits, and conducts legal defense activities in court using generative AI, with an emotion engine that recognizes the user's emotions. This system consists of a server, a terminal, a user, and an emotion engine.

[1153] Data collection and storage

[1154] The server periodically collects past civil and family law case data from legal-related public data sources and open databases. To do this, it uses web scraping tools (e.g., Beautiful Soup, Scrapy) and APIs. The collected data is normalized and checked for quality before being stored in a database (e.g., MySQL, PostgreSQL, MongoDB). The data is managed by removing duplicate data and filling in missing values, and an index is created to enable search and reference as needed.

[1155] Litigation information entry and analysis

[1156] Users use a device (e.g., PC, tablet, or smartphone) to enter information related to their legal case. Specifically, they fill out a form on the device, including the nature of the case, the desired outcome, and supporting documents. During this process, the emotion engine uses the device's camera and microphone to analyze the user's facial expressions and voice to recognize their emotional state in real time. Emotional states include stress levels, anxiety, and joy.

[1157] Submitting Litigation Information

[1158] The terminal transmits the entered litigation information and the emotion data recognized by the emotion engine to the server. This transmission is performed using encrypted communication (e.g., HTTPS).

[1159] Searching for similar cases and calculating the winning rate

[1160] The server searches for similar past cases in its database based on the received user's lawsuit information and emotional data. To do this, it uses natural language processing (NLP) algorithms (e.g., BERT, Word2Vec) and similarity calculation algorithms (e.g., Cosine Similarity). Based on the data of similar cases found, it uses machine learning models (e.g., logistic regression, random forest) to calculate the success rate of the lawsuit. The user's emotional state is also reflected in the prediction model to improve the accuracy of the success rate.

[1161] Generate and provide a report of the results

[1162] The server automatically generates a report containing the winning probability calculation results and details of related similar cases. This report includes detailed explanations to maintain transparency of the calculation process and also serves the purpose of reducing the mental burden on the user. The generated report is displayed to the user via the terminal, and the user can operate it intuitively using the user interface.

[1163] Generative AI advocacy

[1164] When legal amendments are made, the server collects data on court records and lawyers' legal activities and trains this data on a generative AI model (e.g., GPT-3, BERT). This allows the generative AI to acquire legal skills to support litigation. When a user requests the generative AI to act as a legal representative, they re-enter the necessary legal information using their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the generative AI. The server then formulates an optimal legal strategy based on the received information and learning data. Once the lawsuit begins, the generative AI actually acts as a legal representative in court, providing real-time advice to the user and supporting the trial progress.

[1165] Specific examples

[1166] For example, suppose a user is considering divorce proceedings and is battling over asset distribution and child custody. The user uses a device to input the details of the lawsuit and supporting documents. At this time, the emotion engine recognizes emotions such as stress and anxiety from the user's facial expressions and voice. The server receives this information and searches its database for similar past divorce lawsuit cases. Based on the search results, taking emotions into account, the server calculates a 70% chance of success and provides a report to the user.

[1167] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the Generator AI defend them. The emotion engine again recognizes the user's emotions and reflects them in the Generator AI. The Generator AI then conducts the defense in court and provides real-time advice to the user. Finally, if the Generator AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[1168] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment where ordinary households can easily receive legal support. Furthermore, by providing support that takes the user's emotions into consideration, it is possible to make more appropriate legal decisions while reducing the mental burden.

[1169] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1170] Step 1: Data collection and database update

[1171] The server collects past civil and family law case data from legal-related public data sources and open databases. Specifically, data is periodically retrieved using web scraping tools (e.g., Beautiful Soup, Scrapy) and APIs. The collected data is then normalized and its format is checked for quality. This step also involves removing duplicate data and imputing missing values. The input data is obtained from public data sources, and the output data is normalized case data. The data is then stored in a database (e.g., MySQL, PostgreSQL, MongoDB). An index is built for the stored data to enable efficient access for later search and analysis.

[1172] Step 2: Enter case information and recognize emotional state

[1173] Users use a device (e.g., PC, tablet, or smartphone) to enter information related to their legal case. Specifically, they fill out an input form with the details of the lawsuit, the desired outcome, and supporting documents. At the same time, the emotion engine uses the device's camera and microphone to analyze the user's facial expressions and voice to recognize their emotional state in real time. Emotional states include stress levels, anxiety, and joy. The input data are the lawsuit information and emotion data provided by the user, and the output data are the analyzed lawsuit information and emotion state.

[1174] Step 3: Submit your case information

[1175] The terminal transmits the entered legal information and the emotion data recognized by the emotion engine to the server. This transmission is performed using encrypted communication (e.g., HTTPS). The input data is the legal information and emotion data, and the output data is the data securely transmitted to the server.

[1176] Step 4: Find similar cases

[1177] The server searches for similar past cases in the database based on the received user's lawsuit information and emotion data. To do this, it uses natural language processing (NLP) algorithms (e.g., BERT, Word2Vec) and similarity calculation algorithms (e.g., Cosine Similarity). The input data are the received lawsuit information and emotion data, and the output data are the searched similar cases.

[1178] Step 5: Calculating Win Rate

[1179] The server uses data extracted from similar cases to calculate the probability of winning a lawsuit using a machine learning model (e.g., logistic regression, random forest). This step also takes into account the user's emotional data. For example, if the user is very nervous, the predictive model will compensate for that effect. The input data are similar cases and emotional data, and the output data is the calculated probability of winning a lawsuit.

[1180] Step 6: Generate reports

[1181] The server automatically generates a report containing the winning probability calculation results and details of related similar cases. This report includes detailed explanations to maintain transparency of the calculation process and also serves the purpose of reducing the mental burden on the user. The input data are the winning probability calculation results and detailed data of similar cases, and the output data is the generated report.

[1182] Step 7: Provide a report

[1183] The terminal displays the generated report to the user. The user interface is designed to be intuitive. The user views the report through the terminal and decides whether to proceed with the lawsuit. Feedback from the emotion engine is also provided here. The input data is the generated report, and the output data is the report presentation and feedback to the user.

[1184] Step 8: Litigation training of generative AI

[1185] When a legal amendment is made, the server collects data on court records and lawyers' legal activities and trains this data into a generative AI model (e.g., GPT-3, BERT). The input data is the revised court records and legal activities data, and the output data is the model data trained by the generative AI.

[1186] Step 9: Defending the generative AI

[1187] When a user requests a defense from the generative AI, they re-enter the necessary legal information into their device and send the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the generative AI. The server then formulates an optimal defense strategy based on the received information and learning data. The input data is the re-entered legal information and emotion data, and the output data is the defense strategy created by the generative AI.

[1188] Step 10: Real-time advocacy

[1189] Once the lawsuit begins, the generative AI will actually conduct the defense in court. It will provide real-time advice to the user and support the trial, including suggesting when to present evidence and what questions to ask. It will continuously monitor the user's emotional state using an emotion engine, and the generative AI will adjust its strategy as needed. The input data is the user's real-time emotional state and the status of the trial, and the output data is real-time advice and support from the generative AI.

[1190] (Application example 2)

[1191] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1192] While conventional litigation support systems can calculate the probability of winning based on past case data, they do not take into account the user's emotional state and lack support for mental stress and anxiety. Furthermore, because they do not provide user support or emotional monitoring in real time during legal proceedings, they do not alleviate the psychological burden on users in the courtroom. This hinders user decision-making and effective legal proceedings during legal proceedings.

[1193] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1194] In this invention, the server includes means for collecting and storing case data from past civil and family law cases in a database; means for receiving case information entered by a user and searching the database for similar past cases; means for calculating the success rate of the case based on the searched similar cases and providing the result to the user; means for recognizing the user's emotional state using an emotion engine and more accurately calculating the success rate of the case taking the emotional state into account; and means for generating and providing a report that takes the user's emotions into account. This enables accurate calculation of the success rate of the case taking the user's emotional state into account and also provides psychological support for making litigation decisions. Furthermore, by monitoring the user's emotions and providing mental support during real-time legal proceedings, the server can reduce the user's psychological burden and support the effective progress of the case.

[1195] A "civil lawsuit" is a lawsuit to resolve legal disputes between private citizens, where individuals or entities dispute each other's rights and obligations.

[1196] A "family litigation" is a judicial proceeding to resolve disputes about domestic issues and relationships.

[1197] "Case law data" is a record of past decisions made by courts on legal issues.

[1198] A "database" is a system for efficiently searching, storing, and managing large amounts of organized data.

[1199] "Similar cases" are past court decisions that have similar facts and legal issues to those in the current lawsuit.

[1200] A "success rate" is the percentage of chances of success in a lawsuit.

[1201] "Generative AI" is artificial intelligence that generates and analyzes new information based on machine learning and deep learning.

[1202] An "emotion engine" is a system that recognizes and analyzes human emotions from voice, facial expressions, context, etc.

[1203] A "report" is a report that summarizes analysis results and information.

[1204] A "trial record" is a record of all documents and evidence arising in the course of a trial.

[1205] "Advocacy" refers to the work and actions of a lawyer to assist and defend a client under the law.

[1206] A "user interface" is a screen or operating means that allows a user to interact with a system or application.

[1207] "Mental support" refers to services that help users to alleviate and support their mental and emotional anxieties and difficulties.

[1208] This invention is a system that calculates the success rate of lawsuits based on past legal precedent data, and combines advocacy activities using generative AI with user support using an emotion engine. This system consists of a server, terminals, users, and an emotion engine.

[1209] Data collection and storage

[1210] The server collects historical civil and family law case data from legal-related public data sources and open databases using web scraping and APIs, normalizes the data, and stores it in a database that is managed for efficient searching and referencing.

[1211] Entering lawsuit information and calculating win rate

[1212] Using the terminal, the user inputs information related to their legal case. During this information input process, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The emotional state and legal case information are then sent to the server.

[1213] The server searches a database based on the received lawsuit information and sentiment data to extract similar past lawsuits. This search uses natural language processing (NLP) algorithms and similarity calculations. Using the analysis results of the extracted similar cases and sentiment data, a machine learning model (e.g., logistic regression or random forest) is used to calculate the winning rate.

[1214] Calculating winning rates and providing results

[1215] The server uses the emotion engine to take into account the user's emotional state and more accurately calculate the success rate of the lawsuit. A report is generated containing the calculation results and details of related similar cases. The report is displayed to the user through the user interface on their device, allowing them to view the success rate, similar cases, and specific advice. Based on the report and advice provided, the user decides whether to proceed with the lawsuit. Feedback from the emotion engine can also be used to help make legal decisions.

[1216] Generative AI advocacy

[1217] When legal amendments are made or new case data is added, the server collects court records and data on lawyers' legal activities and trains the generative AI. Based on the learned data, the generative AI acquires legal skills to support litigation.

[1218] When a user requests the Generative AI to defend their case, they re-enter the necessary legal information into their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects that information in the Generative AI. The server then formulates an optimal defense strategy based on the received information and learning data. Once the lawsuit begins, the Generative AI will conduct defense activities in court and provide advice to the user in real time. The emotion engine continuously monitors the user's emotional state, and the Generative AI will adjust its strategy as needed.

[1219] Specific examples

[1220] For example, if a company reports a data breach incident, they can use the AI ​​security assistant application to input details of the incident. The emotion engine analyzes the stress and anxiety of the person in charge and compares it with similar cases in the database to conduct a risk assessment. The results are provided as a report, and any necessary mental support is also provided.

[1221] Example prompt sentence:

[1222] "Important data was leaked by an internal employee. As a countermeasure, we are considering strengthening log management and providing employee training."

[1223] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1224] Step 1:

[1225] The server collects past civil and family law case data from legal public data sources and open databases, using web scraping and APIs, and stores it in a database. The data is then normalized and stored in the database.

[1226] Input: Public legal data sources

[1227] Output: Normalized case data

[1228] Step 2:

[1229] The user uses the device to input information related to their legal case (e.g., the details of the case, the desired outcome, and supporting documents). As the user inputs information, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state.

[1230] Input: Lawsuit case information, user facial expression and voice data

[1231] Output: Input case information, perceived emotional state

[1232] Step 3:

[1233] The server searches the database based on the received user's lawsuit information and emotion data to extract similar past lawsuits, using natural language processing (NLP) algorithms and similarity calculations.

[1234] Input: User's lawsuit information, emotional data

[1235] Output: Extracted similar lawsuit cases

[1236] Step 4:

[1237] Based on the extracted similar lawsuits and sentiment data, the server uses machine learning models to calculate the probability of winning a lawsuit, using algorithms such as logistic regression and random forests.

[1238] Input: Similar lawsuit cases, emotional data

[1239] Output: Calculated success rate of lawsuits

[1240] Step 5:

[1241] The server further adjusts the chances of winning the case by taking into account the emotional state of the user as recognized using the emotion engine.

[1242] Inputs: Calculated case win rate, perceived emotional state

[1243] Output: Adjusted lawsuit win rate

[1244] Step 6:

[1245] The server generates a report containing the results of the calculation and details of related similar cases, which can be transmitted to the terminal and viewed by the user through the user interface.

[1246] Input: Adjusted case win rate, details of related similar cases

[1247] Output: Generated report

[1248] Step 7:

[1249] The user decides whether to proceed with the lawsuit based on the report and advice provided through the terminal. During this decision-making process, the emotion engine monitors the user's emotional state and provides feedback.

[1250] Input: Report, Advice

[1251] Output: User's legal decision

[1252] Step 8:

[1253] When legal amendments or new case law data are added, the server collects court records and data on lawyers' legal activities and trains the generation AI.

[1254] Input: Latest court records and advocacy data

[1255] Output: A trained generative AI model

[1256] Step 9:

[1257] When a user requests the AI ​​to represent them, they re-enter the legal information into their device, which is then sent to the server. During this process, the emotion engine recognizes the user's emotions and reflects them in the AI.

[1258] Input: Re-entered case information, perceived emotional state

[1259] Output: Data passed to the generative AI

[1260] Step 10:

[1261] The server uses the trained generative AI model to develop an optimal defense strategy and provides real-time advice to the user during the courtroom. The emotion engine continuously monitors the user's emotional state, and the generative AI adjusts the strategy.

[1262] Input: Generative AI model, real-time user emotion data

[1263] Output: Real-time advice and tailored defense strategies

[1264] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1265] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1266] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1267] [Fourth embodiment]

[1268] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1269] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1270] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1271] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1272] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1273] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1274] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1275] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1276] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1277] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1278] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1279] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1280] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1281] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data of civil and family law lawsuits, and furthermore, uses generative AI to conduct defense activities in court. This system operates in cooperation with three parties: a server, a terminal, and a user.

[1282] Data collection and storage

[1283] The server periodically collects case data for past civil and family law cases from legal-related public data sources and open databases. This data includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[1284] Entering lawsuit information and calculating win rate

[1285] The user uses the terminal to input information related to their legal case, including the specifics of the case, the desired outcome, and supporting documents. This input data is then sent from the terminal to the server.

[1286] The server searches the database based on the received user's lawsuit information to extract similar past cases. This is done using natural language processing (NLP) algorithms and similarity calculations. Based on the analysis results of the extracted similar cases, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the lawsuit. The calculation results are generated in report format and sent to the terminal.

[1287] The user can review the reports provided by the terminal and view details of the winning rate and related cases. Based on this information, the user can decide whether to proceed with the lawsuit.

[1288] Generative AI advocacy

[1289] When legal changes are made, the server collects data on court records and lawyers' legal activities and trains the generative AI. This training data includes minutes, evidence, and legal content. The generative AI acquires legal skills to support litigation based on the learned data.

[1290] When a user requests that the generated AI represent them, they input the necessary information into their device and send a request for the generated AI to the server. The server then uses this information and learning data to formulate an optimal defense strategy. When the generated AI represents its case in court, it provides real-time advice to the user and supports the progress of the trial.

[1291] Users can monitor the AI's defense activities in court and provide instructions as necessary. Once the outcome of the lawsuit is known, a detailed report will be provided via the device, along with advice on future actions.

[1292] Specific examples

[1293] For example, suppose a user is considering divorce proceedings and is fighting over asset division and child custody. The user uses a terminal to enter the details of the lawsuit and supporting documents. The server receives this information and searches its database for similar past divorce cases. Based on the search results, the server calculates a 70% chance of success and provides a report to the user.

[1294] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend their case. The generated AI will study past court records and conduct a defense in court. Finally, if the generated AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[1295] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment in which ordinary households can easily receive legal support.

[1296] The processing flow will be explained below.

[1297] Step 1:

[1298] The server collects historical civil and family law case data from legal public data sources and open databases, using web scraping and APIs to normalize and retrieve the data.

[1299] Step 2:

[1300] The server preprocesses the collected case data, converting it into a standard format before storing it in a database, including deduplication, filling in missing values, and text cleansing.

[1301] Step 3:

[1302] The user uses a terminal to enter information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents). The input form has required and optional fields, and the user must enter all required information.

[1303] Step 4:

[1304] The terminal transmits the entered litigation information to the server, where the transmitted data is encoded and sent using a secure communication method.

[1305] Step 5:

[1306] The server analyzes the received user's lawsuit information and searches for similar past cases in its database, using natural language processing (NLP) algorithms and similarity calculations.

[1307] Step 6:

[1308] The server calculates the probability of winning a lawsuit based on the data of similar cases retrieved, using machine learning models (such as logistic regression and random forests) to take into account necessary variables.

[1309] Step 7:

[1310] The server generates a report detailing the winning probability calculation and related similar cases, including a detailed explanation of the calculation process to ensure transparency.

[1311] Step 8:

[1312] The server sends the generated reports to the terminal, which are presented in a visually easy-to-understand format (graphs, tables, text).

[1313] Step 9:

[1314] The terminal displays the submitted report to the user through a user interface, where the user can view the win rate, similar cases, and specific advice.

[1315] Step 10:

[1316] Users can decide whether to proceed with a lawsuit based on the reports and advice provided, and can also ask additional questions or complete the data.

[1317] Step 11:

[1318] When legal changes are made, the server collects data on court records and lawyers' legal activities, including minutes, evidence, and content of legal arguments, and organizes it into a learning dataset.

[1319] Step 12:

[1320] The server trains the generative AI using data based on court records, using deep learning techniques to train the AI ​​to acquire effective advocacy skills.

[1321] Step 13:

[1322] To request the generated AI to defend the case, the user re-enters the necessary legal information into the device and submits the request.

[1323] Step 14:

[1324] The server receives a request for legal representation from the user and formulates an optimal legal representation strategy based on the learning data, and if necessary, communicates with the user.

[1325] Step 15:

[1326] Once litigation begins, the generative AI will actually act as a defense in court, providing real-time advice to the user and supporting the proceedings, including suggesting when to present evidence and what questions to ask.

[1327] Step 16:

[1328] After the verdict is reached, the server analyzes the results and generates a detailed report that is sent to the device, including the outcome of the verdict and advice on future actions.

[1329] Step 17:

[1330] The user will use the provided report to determine any necessary follow-up actions (eg, appeal decisions, etc.).

[1331] The above are the specific processing steps of the entire system.

[1332] Example 1

[1333] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1334] Existing legal support systems have difficulty calculating an appropriate win rate for each user's legal information, or providing real-time support for legal activities based on past cases. Furthermore, to efficiently support legal activities in court, it is necessary to analyze a huge amount of past court records and lawyers' legal activity data, and use generative AI to formulate optimal legal strategies.

[1335] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1336] In this invention, the server includes means for collecting past litigation case data from legal-related public data sources and open databases and storing it in a database, means for receiving litigation information entered by a user and searching the database for similar past cases using a natural language processing algorithm, means for analyzing the searched similar cases and calculating the success rate of the litigation using a machine learning model and providing the results to the user in the form of a report, and means for the user to use a terminal for inputting litigation information and evidential materials. This allows users to calculate the success rate of their own litigation cases with high accuracy and to develop optimal defense strategies based on similar past cases.

[1337] "Legal-related public data sources" refers to databases of laws, precedents, court records, etc. that are publicly available from governments and related agencies.

[1338] "Open database" refers to a database provided by a third party that is freely accessible on the Internet.

[1339] "Litigation case data" refers to a collection of data including records of past civil and family law litigation, judgments, related laws and regulations, evidence lists, and background information on the parties involved.

[1340] "Database" refers to an information system for efficiently storing, retrieving, and managing information.

[1341] "Law case information entered by the user" refers to the specific details, desired outcome, supporting documents, etc., of the user's own legal case.

[1342] "Natural language processing algorithm" refers to technology that allows computers to analyze human language and understand its meaning.

[1343] "Similar past cases" refers to past legal cases that have similar characteristics based on specific criteria to the user's legal information.

[1344] A "machine learning model" refers to a statistical algorithm that uses data to learn and make predictions, classifications, and other similar tasks.

[1345] "Generative AI" refers to artificial intelligence that can learn from past data and automatically perform specific tasks.

[1346] A "defense strategy" refers to a plan or method that aims to gain an advantage in a lawsuit based on law and evidence.

[1347] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data for civil and family law lawsuits, and uses generative AI to conduct defense activities in court. This system operates in cooperation with three parties: a server, a terminal, and a user.

[1348] Data collection and storage

[1349] The server periodically collects past civil and family law case data from legal-related public data sources and open databases using curl or API. The collected data is then stored in a database (e.g., MySQL or PostgreSQL) after checking its accuracy. It is also indexed to streamline data search and management.

[1350] Entering lawsuit information and calculating win rate

[1351] The user enters information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents) into a dedicated form on the device. The device temporarily stores the entered data, and once all the data is collected, it sends an HTTP POST request to the server.

[1352] The server uses a natural language processing (NLP) algorithm to search its database for similar past cases based on the received user's lawsuit information. Specifically, it generates a search query and performs a full-text search of the past lawsuit database. Based on the analysis results of the extracted similar cases, it uses a machine learning model (e.g., logistic regression or random forest) to calculate the success rate of the lawsuit.

[1353] The results are formatted into a report and sent from the server to the terminal. The user can then review the report provided by the terminal and refer to the winning rate and details of related cases. Based on this information, the user can decide whether to proceed with the lawsuit.

[1354] Generative AI advocacy

[1355] When legal changes or new legal precedents arise, the server collects court records and lawyers' advocacy data and trains the generative AI using a natural language processing framework (e.g., BERT). This training data includes court records, evidence, and advocacy content.

[1356] When a user requests a defense from the generated AI, they input the necessary information into their device and send a defense request to the server. The server receives this information and provides the necessary data to the generated AI.

[1357] The generative AI will develop an optimal defense strategy based on past learning data and support legal actions in court. It will send advice to the user in real time to support the trial. The user can monitor the generative AI's legal actions in court and give instructions as necessary. Once the outcome of the trial is known, a detailed report will be provided to the user via their device, along with advice on future countermeasures.

[1358] Specific examples

[1359] For example, suppose a user is considering divorce proceedings and is fighting over asset division and child custody. The user uses a terminal to enter the details of the lawsuit and supporting documents. The server receives this information and searches its database for similar past divorce cases. Based on the search results, the server calculates a 70% chance of success and provides a report to the user.

[1360] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend their case. The generated AI will learn from past court records and conduct a defense in court. Finally, if the generated AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[1361] An example prompt is, "I am currently in a divorce proceeding regarding asset division and child custody. How can I calculate my chances of winning based on past case data and use generative AI to defend my case in court?"

[1362] The above is a specific embodiment of the present invention. This system will reduce the cost and time required for litigation, and create an environment in which ordinary households can easily receive legal support.

[1363] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1364] Step 1:

[1365] The server periodically collects past civil and family law case data from legal-related public data sources and open databases via the Internet, using curl and APIs to retrieve the data.

[1366] Input: Data source URL or API endpoint

[1367] Data processing: Formatting the acquired data and checking its accuracy

[1368] Output: A formatted dataset

[1369] Specific example: "curl https: / / example-legal-database.com / api / cases"

[1370] Step 2:

[1371] The server stores the collected data in a database (e.g., MySQL or PostgreSQL) and indexes it for efficient searching.

[1372] Input: Formatted dataset

[1373] Data manipulation: Inserting and indexing data based on the database schema

[1374] Output: Data stored in the database

[1375] Specific example: "INSERT INTO cases_table (case_id, case_type, verdict, laws, evidence, parties_info) VALUES (...)"

[1376] Step 3:

[1377] The user enters the case information into a dedicated form on the terminal.

[1378] Input: Contents of lawsuit, desired outcome, evidence, etc.

[1379] Data calculation: Temporarily save input data

[1380] Output: Temporarily saved dataset

[1381] Specific example: "Enter the details of the lawsuit in text on a web form"

[1382] Step 4:

[1383] The terminal sends the completed input data to the server via an HTTP POST request.

[1384] Input: Completed dataset

[1385] Data processing: Generating the request body

[1386] Output: Request sent to server

[1387] Example: "POST / api / case_submission { case_data: {...}}"

[1388] Step 5:

[1389] Based on the user's lawsuit information received by the server, a natural language processing (NLP) algorithm is used to search for similar past cases.

[1390] Input: User's case information

[1391] Data Computing: Database Querying Using Full-Text Search Algorithms

[1392] Output: List of similar jobs

[1393] Specific example: SELECT FROM cases_table WHERE MATCH (case_text) AGAINST ('User's lawsuit details')

[1394] Step 6:

[1395] The server analyzes data from similar cases and uses machine learning models (such as logistic regression or random forest) to calculate the probability of winning a lawsuit.

[1396] Input: Data for similar cases

[1397] Data Computation: Probability Computation Using Machine Learning Models

[1398] Output: Win rate calculation result

[1399] Specific example: "Calculating probability using a machine learning model (e.g., sklearn's LogisticRegression.predict_proba)"

[1400] Step 7:

[1401] The server formats the winning rate calculation results into a report format and sends it to the terminal.

[1402] Input: Win rate calculation result

[1403] Data processing: Report generation (e.g. HTML or PDF format)

[1404] Output: Report sent to user terminal

[1405] Specific example: "Using a report generation library (e.g. ReportLab)"

[1406] Step 8:

[1407] The user reviews the report on the terminal and decides whether to proceed with the lawsuit.

[1408] Input: Report data

[1409] Data calculation: Understanding and evaluating report content

[1410] Output: Decision on next action

[1411] Specific example: "Display report PDF on browser"

[1412] Step 9:

[1413] The user inputs the necessary information to request legal representation from the generated AI and sends it from the terminal to the server.

[1414] Input: Detailed information about the legal request

[1415] Data processing: Generating request data

[1416] Output: Request sent to server

[1417] Specific example: "Enter legal request information on a web form"

[1418] Step 10:

[1419] The server provides the received request information to the generation AI, which then formulates the optimal defense strategy.

[1420] Input: Legal request information

[1421] Data Computation: Strategy Planning with Generative AI

[1422] Output: Strategic proposal

[1423] Specific example: "API call to pass request information to the generation AI"

[1424] Step 11:

[1425] Generative AI will act as a defense in court and provide real-time advice to users.

[1426] Input: Strategic data generated by generative AI

[1427] Data Computing: Real-time Advice Generation

[1428] Output: Provide advice to the user

[1429] Specific example of operation: "Generative AI sends advice in real time in chat format"

[1430] Step 12:

[1431] The user will monitor the generated AI's defense activities in court and provide instructions as necessary.

[1432] Input: Generative AI advice and on-site situation

[1433] Data Computing: Advice and Situation Assessment

[1434] Output: Sending instructions

[1435] Specific example: "Real-time feedback and instructions on web applications"

[1436] Step 13:

[1437] When the outcome of the lawsuit becomes clear, the server generates a detailed report and sends it to the device, along with advice on future countermeasures.

[1438] Input: Lawsuit outcome data

[1439] Data processing: detailed report generation and analysis of future countermeasures

[1440] Output: Detailed report and further advice to the user

[1441] Specific example: "Generating and sending a final result report"

[1442] (Application example 1)

[1443] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1444] In the past, in civil and family law cases, it was difficult to calculate the success rate of a case based on past case data, and there was a lack of means to provide real-time advice to properly support legal advocacy in court. As a result, in many cases, users found it difficult to make informed decisions, and the quality of legal advocacy was inconsistent.

[1445] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1446] In this invention, the server includes means for collecting case data on past civil and family law litigation cases and storing it in a database, means for receiving litigation information entered by a user and searching the database for similar past cases, means for calculating the success rate of the litigation based on the searched similar cases and providing the result to the user in real time, and means for receiving the litigation information entered by a user via a smartphone and calculating the success rate of the litigation in real time. This allows the user to obtain a highly accurate litigation success rate based on past cases and to obtain the necessary information in real time via their smartphone.

[1447] "Past civil and family law case precedent data" refers to past civil and family law case precedent information collected from public institutions and legal databases.

[1448] "Means of storing in a database" refers to a method or system for systematically organizing collected case law data and storing it so that it can be easily searched and analyzed.

[1449] "Means for receiving litigation information entered by the user and searching for similar past cases" refers to technologies and algorithms for obtaining information about litigation provided by the user and using that information to find similar cases from past case data.

[1450] "Means for calculating the success rate of a lawsuit and providing the result to the user in real time" refers to technology or methods that analyze past case law data and information provided by the user to calculate the likelihood that the outcome of the lawsuit will be favorable to the user and present the result to the user in real time.

[1451] "Means of obtaining necessary information in real time via a smartphone" refers to applications and systems that allow users to receive information provided in real time via a smartphone, which is a mobile device.

[1452] "Means for training generative AI" refers to methods and techniques for collecting data on court records and legal actions and using them to train generative AI models.

[1453] "Means for inputting information to request defense from the generated AI" refers to an interface through which a user can input the information necessary to request defense from the generated AI.

[1454] "Means for generative AI to conduct legal advocacy in court" refers to technologies and systems that enable generative AI to conduct legal advocacy in court based on data learned by the AI ​​and provide legal advice in real time.

[1455] "Means for visually displaying through a user interface" refers to an interface that allows a user to visually check the results of the lawsuit's winning probability calculation and details of similar precedents.

[1456] "Means for providing advice to help users decide whether to proceed with litigation" refers to a system or method for providing information or advice to users that will help them decide whether to proceed with litigation.

[1457] This invention relates to a system that calculates the success rate of lawsuits based on past precedent data for civil and family law cases, and then uses generative AI to conduct defense activities in court. This system operates in cooperation with a server, a terminal, and a user.

[1458] Data collection and storage

[1459] First, the server periodically collects case data on past civil and family law cases from public institutions and legal databases. This data includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties involved. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[1460] Entering lawsuit information and calculating win rate

[1461] Users use a device (such as a smartphone) to enter information related to their legal case. This information includes the specific details of the case, the desired outcome, and supporting documents. The entered data is sent from the device to a server. The server then searches a database based on the user's legal case information and extracts similar past cases. This process uses natural language processing (NLP) algorithms and similarity calculations. Based on the analysis of the extracted similar cases, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the case. The calculation results are generated in the form of a report and sent to the device.

[1462] Generative AI advocacy

[1463] When legal amendments are made, the server collects data on court records and lawyers' legal activities and trains the generated AI. This training data includes minutes, evidentiary documents, and legal content. The generated AI uses the trained data to acquire legal skills to support litigation. When a user requests the generated AI to act as a legal representative, they enter the necessary information on their device and send a request for legal representation to the server. The server then formulates the optimal legal strategy based on this information and the training data. When the generated AI acts as a legal representative in court, it sends advice to the user in real time to support the progress of the trial. The user can check the generated AI's legal activities in court and issue instructions as necessary. Once the outcome of the lawsuit is known, a detailed report is provided via the device, along with advice on future countermeasures.

[1464] User Interface

[1465] The calculation results of the winning rate and details of similar cases sent from the server are visually displayed through the user interface, making it easier for users to intuitively understand the lawsuit information. The same interface also provides advice to help users decide whether to proceed with the lawsuit.

[1466] Specific examples

[1467] For example, suppose a user is considering divorce proceedings and is fighting over asset distribution and child custody. The user uses a terminal to input the details of the lawsuit and supporting documents. The server receives this information and searches a database for similar past divorce cases. Based on the search results, it calculates a 70% chance of success and provides a report to the user. If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the generated AI defend the case. The generated AI has learned from past court records and will defend the case in court. Finally, if the generated AI's defense results in a favorable verdict, the user reviews the results and decides what subsequent actions are required.

[1468] Examples of prompts for generative AI models

[1469] If you are considering divorce proceedings and are disputing asset distribution and child custody, please analyze past legal precedents related to this type of case and propose legal action that focuses on the fairness of asset distribution and the best interests of the children.

[1470] The above is a specific embodiment of the present invention.

[1471] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1472] Step 1: Collect and store data

[1473] The server periodically collects case data for past civil and family law cases from legal-related public data sources and open databases. The data input includes the type of case, the content of the judgment, relevant laws and regulations, a list of evidence, and background information on the parties. This data is stored in a database and managed for easy searching and reference.

[1474] Step 2: Enter your case information

[1475] Users use their devices (smartphones) to input specific information about their legal case—the nature of the case, the desired outcome, evidence, etc.—and this input data is sent from the device to the server.

[1476] Step 3: Find similar cases

[1477] The server searches the database based on the user's lawsuit information received in step 2 to extract similar past cases. This uses natural language processing (NLP) algorithms and similarity calculations. Specifically, the entered lawsuit content is tokenized, vectorized using techniques such as TF-IDF, and compared with similar cases in the database.

[1478] Step 4: Calculating the probability of success of your lawsuit

[1479] The server uses a machine learning model (such as logistic regression or random forest) to calculate the success rate of the lawsuit based on the analysis results of similar cases extracted in step 3. To do this, the input data and statistical data of similar cases are plotted and input into the model to predict the success rate. The calculation results are generated in report format and sent to the terminal.

[1480] Step 5: Litigation Support Report

[1481] The terminal receives the report sent from the server in step 4. The report includes details of similar cases along with the calculated win rate. The user can view this information and use it to decide whether to proceed with the lawsuit.

[1482] Step 6: Generative AI learns to advocate

[1483] When legal changes occur, the server collects court records and data on lawyers' legal activities and trains the AI. This training data includes minutes, evidence, and legal content. The AI ​​uses the collected data to acquire legal skills.

[1484] Step 7: Enter your request for legal representation

[1485] When a user requests a generated AI to defend them, they input the necessary information into their device and send the request to the server, including the specific details of the lawsuit and the key points of the defense they are seeking.

[1486] Step 8: Developing a defense strategy

[1487] The server formulates an optimal defense strategy based on the user information received in step 7 and the data learned by the generating AI. The generating AI utilizes the learned data to provide real-time advice to the user for advocacy in court.

[1488] Step 9: Real-time legal support

[1489] As users pursue their cases in court, the generative AI provides real-time support for their defense, providing advice on how to present evidence and what to say. Users can use this advice to make decisions in court.

[1490] Step 10: Reporting the outcome of the case

[1491] Once the case is over, the device receives a detailed report, including the outcome of the case and the effectiveness of the advice provided by the AI. The user can use this information to decide on future actions.

[1492] These are the specific processing steps of the system. Based on the specific software and algorithms used in each step, we can clearly understand the entire process from data input to output.

[1493] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1494] This invention combines a system that calculates the success rate of lawsuits based on past precedent data from civil and family law lawsuits, and conducts legal defense activities in court using generative AI, with an emotion engine that recognizes the user's emotions. This system consists of a server, a terminal, a user, and an emotion engine.

[1495] Data collection and storage

[1496] The server periodically collects past civil and family law case data from legal-related public data sources and open databases. This data is obtained by normalizing it using web scraping and APIs. The collected data is stored in a database and managed so that it can be searched and referenced as needed.

[1497] Entering lawsuit information and calculating win rate

[1498] The user uses the terminal to input information related to their legal case (e.g., the lawsuit content, desired outcome, and supporting documents). During the process of inputting this information, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The emotion recognition results are sent to the server along with the input information.

[1499] The server searches the database based on the received user's lawsuit information and sentiment data to extract similar past cases. This is done using natural language processing (NLP) algorithms and similarity calculations. Based on the analysis results of the extracted similar cases and sentiment data, a machine learning model (e.g., logistic regression or random forest) is used to calculate the success rate of the lawsuit.

[1500] Calculating winning rates and providing results

[1501] The server uses an emotion engine to take into account the user's recognized emotional state to more accurately calculate the probability of winning the lawsuit. A report is generated containing the calculation results and details of related similar cases. The report includes detailed explanations to ensure transparency of the calculation process and is presented to the user in a manner that minimizes mental burden.

[1502] The terminal displays the submitted report to the user through a user interface. The user can view the win rate, similar cases, and specific advice. Based on the provided report and advice, the user decides whether to proceed with the lawsuit. Feedback from the emotion engine can also be used to help make a lawsuit decision.

[1503] Generative AI advocacy

[1504] When legal changes are made, the server collects data on court records and lawyers' legal activities and trains the generative AI on this data. This data includes minutes, evidence, and legal content, and is compiled as a training dataset. The generative AI uses the learned data to acquire legal skills to support litigation.

[1505] When a user requests the AI ​​to represent them, they re-enter the necessary legal information into their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the AI. The server then formulates the optimal defense strategy based on the received information and learning data.

[1506] Once litigation begins, the generative AI will actually conduct the defense in court, providing real-time advice to the user and supporting the trial, including suggesting when to present evidence and what questions to ask. Using an emotion engine, the generative AI will continuously monitor the user's emotional state and adjust its strategy as needed.

[1507] Specific examples

[1508] For example, suppose a user is considering divorce proceedings and is battling over asset distribution and child custody. The user uses a device to input the details of the lawsuit and supporting documents. At this time, the emotion engine recognizes emotions such as stress and anxiety from the user's facial expressions and voice. The server receives this information and searches its database for similar past divorce lawsuit cases. Based on the search results, taking emotions into account, the server calculates a 70% chance of success and provides a report to the user.

[1509] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the Generator AI defend them. The emotion engine again recognizes the user's emotions and reflects them in the Generator AI. The Generator AI then conducts the defense in court and provides real-time advice to the user. Finally, if the Generator AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[1510] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment where ordinary households can easily receive legal support. Furthermore, by providing support that takes the user's emotions into consideration, it is possible to make more appropriate legal decisions while reducing the mental burden.

[1511] The processing flow will be explained below.

[1512] Step 1:

[1513] The server collects historical civil and family law case data from legal public data sources and open databases, using web scraping and APIs to normalize and retrieve the data.

[1514] Step 2:

[1515] The server preprocesses the collected case data, converting it into a standard format before storing it in a database, including deduplication, filling in missing values, and text cleansing.

[1516] Step 3:

[1517] Users use a terminal to input information related to their legal case (e.g., the nature of the case, the desired outcome, and supporting documents). During this input process, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state.

[1518] Step 4:

[1519] The terminal transmits the entered lawsuit information and the user's emotion data recognized by the emotion engine to the server. The transmitted data is encoded and sent using a secure communication method.

[1520] Step 5:

[1521] The server analyzes the received user's lawsuit information and sentiment data and searches for similar past cases in its database, using natural language processing (NLP) algorithms and similarity calculations.

[1522] Step 6:

[1523] The server calculates the probability of winning a lawsuit based on the data of similar cases it has searched for, using machine learning models (such as logistic regression or random forest) and taking into account necessary variables.

[1524] Step 7:

[1525] The server adjusts the probability of winning a lawsuit by taking into account the user's emotional state, which is recognized using an emotion engine. For example, if the user is in a high stress state, it is taken into account as a risk factor.

[1526] Step 8:

[1527] The server generates a report detailing the winning probability calculation and related similar cases, including detailed explanations to ensure transparency of the calculation process.

[1528] Step 9:

[1529] The server sends the generated reports to the terminal, which are presented in a visually easy-to-understand format (graphs, tables, text).

[1530] Step 10:

[1531] The terminal displays the submitted report to the user through a user interface, where the user can view the win rate, similar cases, and specific advice.

[1532] Step 11:

[1533] Users can decide whether to proceed with a lawsuit based on the reports and advice provided, and feedback from an emotion engine can also be used to help with the lawsuit decision.

[1534] Step 12:

[1535] When legal changes are made, the server collects data on past court records and lawyers' legal activities, including minutes, evidence, and content of arguments, and uses this data as training data for the generative AI.

[1536] Step 13:

[1537] The server trains the generative AI using data based on court records, using deep learning techniques to train the AI ​​to acquire effective advocacy skills.

[1538] Step 14:

[1539] To request the AI ​​to defend the case, the user re-enters the necessary legal information into the device and submits the request. At this time, the emotion engine again recognizes the user's emotions and reflects them in the AI.

[1540] Step 15:

[1541] The server receives a request for legal representation from the user and formulates an optimal legal representation strategy based on the learning data, and if necessary, communicates with the user.

[1542] Step 16:

[1543] Once litigation begins, the generative AI will actually act as a defense in court, providing real-time advice to the user and supporting the proceedings, including suggesting when to present evidence and what questions to ask.

[1544] Step 17:

[1545] During courtroom advocacy, the generative AI uses an emotion engine to continuously monitor the user's emotional state and adjust its strategy as needed.

[1546] Step 18:

[1547] After the verdict is reached, the server analyzes the results and generates a detailed report that is sent to the device, including the outcome of the verdict and advice on future actions.

[1548] Step 19:

[1549] The user will use the provided report to determine any necessary follow-up actions (eg, appeal decisions, etc.).

[1550] The above are the specific processing steps of the entire system.

[1551] Example 2

[1552] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1553] Conventional litigation support systems could calculate the probability of winning a lawsuit based on past case data, but they were unable to take the user's emotional state into account, resulting in low accuracy in the calculation of the win rate. Furthermore, they were limited in their use of court records and data on lawyers' legal activities, resulting in insufficient courtroom advocacy by the generated AI. Furthermore, they required time and effort to input legal information, and lacked a means to visually present detailed case information. This made it difficult for users to receive appropriate legal support and placed a significant mental burden on them.

[1554] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1555] In this invention, the server includes means for collecting case data on past civil and family law lawsuits and storing it in a database, means for receiving lawsuit information entered by a user and searching the database for similar past cases, means for calculating the success rate of the lawsuit based on the searched similar cases and providing the result to the user, means for analyzing the user's facial expressions and voice and recognizing their emotional state, and means for calculating the success rate by reflecting the emotional state in the calculation means. This enables highly accurate calculation of the success rate that takes the user's emotional state into account, enabling appropriate defense activities in court using generative AI, and providing comprehensive legal support and reducing the mental burden for the user.

[1556] "Past civil and family law case data" refers to a collection of past court decisions and related documents relating to civil and family law cases.

[1557] A "database" is a data management system that systematically stores collected case law data and related information and allows for searching and inquiry.

[1558] "User" refers to an individual or company that utilizes the system to enter information about a lawsuit and receive assistance.

[1559] "Lawsuit information" is information related to the lawsuit entered by the user, including the details of the lawsuit, the desired outcome, and supporting documents.

[1560] "Similar past cases" refers to past precedents that have elements or outcomes similar to the current litigation case.

[1561] "Win rate" is the percentage of the likelihood of success in a lawsuit, calculated based on past similar cases and current litigation information.

[1562] "Analyzing facial expressions and voice to recognize emotional states" refers to the process of reading the user's emotions at that time by analyzing their facial expressions and tone of voice.

[1563] "Generative AI" refers to an artificial intelligence model that learns from past data and automatically responds and makes suggestions to new situations.

[1564] "Courtroom advocacy" refers to a series of activities carried out by lawyers and generative AI in court, such as advocacy, advice, presenting evidence, and proposing questions.

[1565] A "user interface" is a screen or mechanism that allows a user to operate a system, input information, and view results.

[1566] This invention combines a system that calculates the success rate of lawsuits based on past precedent data from civil and family law lawsuits, and conducts legal defense activities in court using generative AI, with an emotion engine that recognizes the user's emotions. This system consists of a server, a terminal, a user, and an emotion engine.

[1567] Data collection and storage

[1568] The server periodically collects past civil and family law case data from legal-related public data sources and open databases. To do this, it uses web scraping tools (e.g., Beautiful Soup, Scrapy) and APIs. The collected data is normalized and checked for quality before being stored in a database (e.g., MySQL, PostgreSQL, MongoDB). The data is managed by removing duplicate data and filling in missing values, and an index is created to enable search and reference as needed.

[1569] Litigation information entry and analysis

[1570] Users use a device (e.g., PC, tablet, or smartphone) to enter information related to their legal case. Specifically, they fill out a form on the device, including the nature of the case, the desired outcome, and supporting documents. During this process, the emotion engine uses the device's camera and microphone to analyze the user's facial expressions and voice to recognize their emotional state in real time. Emotional states include stress levels, anxiety, and joy.

[1571] Submitting Litigation Information

[1572] The terminal transmits the entered litigation information and the emotion data recognized by the emotion engine to the server. This transmission is performed using encrypted communication (e.g., HTTPS).

[1573] Searching for similar cases and calculating the winning rate

[1574] The server searches for similar past cases in its database based on the received user's lawsuit information and emotional data. To do this, it uses natural language processing (NLP) algorithms (e.g., BERT, Word2Vec) and similarity calculation algorithms (e.g., Cosine Similarity). Based on the data of similar cases found, it uses machine learning models (e.g., logistic regression, random forest) to calculate the success rate of the lawsuit. The user's emotional state is also reflected in the prediction model to improve the accuracy of the success rate.

[1575] Generate and provide a report of the results

[1576] The server automatically generates a report containing the winning probability calculation results and details of related similar cases. This report includes detailed explanations to maintain transparency of the calculation process and also serves the purpose of reducing the mental burden on the user. The generated report is displayed to the user via the terminal, and the user can operate it intuitively using the user interface.

[1577] Generative AI advocacy

[1578] When legal amendments are made, the server collects data on court records and lawyers' legal activities and trains this data on a generative AI model (e.g., GPT-3, BERT). This allows the generative AI to acquire legal skills to support litigation. When a user requests the generative AI to act as a legal representative, they re-enter the necessary legal information using their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the generative AI. The server then formulates an optimal legal strategy based on the received information and learning data. Once the lawsuit begins, the generative AI actually acts as a legal representative in court, providing real-time advice to the user and supporting the trial progress.

[1579] Specific examples

[1580] For example, suppose a user is considering divorce proceedings and is battling over asset distribution and child custody. The user uses a device to input the details of the lawsuit and supporting documents. At this time, the emotion engine recognizes emotions such as stress and anxiety from the user's facial expressions and voice. The server receives this information and searches its database for similar past divorce lawsuit cases. Based on the search results, taking emotions into account, the server calculates a 70% chance of success and provides a report to the user.

[1581] If the user decides to proceed with the lawsuit, they enter additional information and submit a request to have the Generator AI defend them. The emotion engine again recognizes the user's emotions and reflects them in the Generator AI. The Generator AI then conducts the defense in court and provides real-time advice to the user. Finally, if the Generator AI's defense results in a favorable verdict, the user can review the results and decide what follow-up actions are required.

[1582] The above is a specific embodiment of the present invention. This system can reduce the cost and time required for litigation and create an environment where ordinary households can easily receive legal support. Furthermore, by providing support that takes the user's emotions into consideration, it is possible to make more appropriate legal decisions while reducing the mental burden.

[1583] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1584] Step 1: Data collection and database update

[1585] The server collects past civil and family law case data from legal-related public data sources and open databases. Specifically, data is periodically retrieved using web scraping tools (e.g., Beautiful Soup, Scrapy) and APIs. The collected data is then normalized and its format is checked for quality. This step also involves removing duplicate data and imputing missing values. The input data is obtained from public data sources, and the output data is normalized case data. The data is then stored in a database (e.g., MySQL, PostgreSQL, MongoDB). An index is built for the stored data to enable efficient access for later search and analysis.

[1586] Step 2: Enter case information and recognize emotional state

[1587] Users use a device (e.g., PC, tablet, or smartphone) to enter information related to their legal case. Specifically, they fill out an input form with the details of the lawsuit, the desired outcome, and supporting documents. At the same time, the emotion engine uses the device's camera and microphone to analyze the user's facial expressions and voice to recognize their emotional state in real time. Emotional states include stress levels, anxiety, and joy. The input data are the lawsuit information and emotion data provided by the user, and the output data are the analyzed lawsuit information and emotion state.

[1588] Step 3: Submit your case information

[1589] The terminal transmits the entered legal information and the emotion data recognized by the emotion engine to the server. This transmission is performed using encrypted communication (e.g., HTTPS). The input data is the legal information and emotion data, and the output data is the data securely transmitted to the server.

[1590] Step 4: Find similar cases

[1591] The server searches for similar past cases in the database based on the received user's lawsuit information and emotion data. To do this, it uses natural language processing (NLP) algorithms (e.g., BERT, Word2Vec) and similarity calculation algorithms (e.g., Cosine Similarity). The input data are the received lawsuit information and emotion data, and the output data are the searched similar cases.

[1592] Step 5: Calculating Win Rate

[1593] The server uses data extracted from similar cases to calculate the probability of winning a lawsuit using a machine learning model (e.g., logistic regression, random forest). This step also takes into account the user's emotional data. For example, if the user is very nervous, the predictive model will compensate for that effect. The input data are similar cases and emotional data, and the output data is the calculated probability of winning a lawsuit.

[1594] Step 6: Generate reports

[1595] The server automatically generates a report containing the winning probability calculation results and details of related similar cases. This report includes detailed explanations to maintain transparency of the calculation process and also serves the purpose of reducing the mental burden on the user. The input data are the winning probability calculation results and detailed data of similar cases, and the output data is the generated report.

[1596] Step 7: Provide a report

[1597] The terminal displays the generated report to the user. The user interface is designed to be intuitive. The user views the report through the terminal and decides whether to proceed with the lawsuit. Feedback from the emotion engine is also provided here. The input data is the generated report, and the output data is the report presentation and feedback to the user.

[1598] Step 8: Litigation training of generative AI

[1599] When a legal amendment is made, the server collects data on court records and lawyers' legal activities and trains this data into a generative AI model (e.g., GPT-3, BERT). The input data is the revised court records and legal activities data, and the output data is the model data trained by the generative AI.

[1600] Step 9: Defending the generative AI

[1601] When a user requests a defense from the generative AI, they re-enter the necessary legal information into their device and send the request. During this process, the emotion engine recognizes the user's emotions and reflects them in the generative AI. The server then formulates an optimal defense strategy based on the received information and learning data. The input data is the re-entered legal information and emotion data, and the output data is the defense strategy created by the generative AI.

[1602] Step 10: Real-time advocacy

[1603] Once the lawsuit begins, the generative AI will actually conduct the defense in court. It will provide real-time advice to the user and support the trial, including suggesting when to present evidence and what questions to ask. It will continuously monitor the user's emotional state using an emotion engine, and the generative AI will adjust its strategy as needed. The input data is the user's real-time emotional state and the status of the trial, and the output data is real-time advice and support from the generative AI.

[1604] (Application example 2)

[1605] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1606] While conventional litigation support systems can calculate the probability of winning based on past case data, they do not take into account the user's emotional state and lack support for mental stress and anxiety. Furthermore, because they do not provide user support or emotional monitoring in real time during legal proceedings, they do not alleviate the psychological burden on users in the courtroom. This hinders user decision-making and effective legal proceedings during legal proceedings.

[1607] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1608] In this invention, the server includes means for collecting and storing case data from past civil and family law cases in a database; means for receiving case information entered by a user and searching the database for similar past cases; means for calculating the success rate of the case based on the searched similar cases and providing the result to the user; means for recognizing the user's emotional state using an emotion engine and more accurately calculating the success rate of the case taking the emotional state into account; and means for generating and providing a report that takes the user's emotions into account. This enables accurate calculation of the success rate of the case taking the user's emotional state into account and also provides psychological support for making litigation decisions. Furthermore, by monitoring the user's emotions and providing mental support during real-time legal proceedings, the server can reduce the user's psychological burden and support the effective progress of the case.

[1609] A "civil lawsuit" is a lawsuit to resolve legal disputes between private citizens, where individuals or entities dispute each other's rights and obligations.

[1610] A "family litigation" is a judicial proceeding to resolve disputes about domestic issues and relationships.

[1611] "Case law data" is a record of past decisions made by courts on legal issues.

[1612] A "database" is a system for efficiently searching, storing, and managing large amounts of organized data.

[1613] "Similar cases" are past court decisions that have similar facts and legal issues to those in the current lawsuit.

[1614] A "success rate" is the percentage of chances of success in a lawsuit.

[1615] "Generative AI" is artificial intelligence that generates and analyzes new information based on machine learning and deep learning.

[1616] An "emotion engine" is a system that recognizes and analyzes human emotions from voice, facial expressions, context, etc.

[1617] A "report" is a report that summarizes analysis results and information.

[1618] A "trial record" is a record of all documents and evidence arising in the course of a trial.

[1619] "Advocacy" refers to the work and actions of a lawyer to assist and defend a client under the law.

[1620] A "user interface" is a screen or operating means that allows a user to interact with a system or application.

[1621] "Mental support" refers to services that help users to alleviate and support their mental and emotional anxieties and difficulties.

[1622] This invention is a system that calculates the success rate of lawsuits based on past legal precedent data, and combines advocacy activities using generative AI with user support using an emotion engine. This system consists of a server, terminals, users, and an emotion engine.

[1623] Data collection and storage

[1624] The server collects historical civil and family law case data from legal-related public data sources and open databases using web scraping and APIs, normalizes the data, and stores it in a database that is managed for efficient searching and referencing.

[1625] Entering lawsuit information and calculating win rate

[1626] Using the terminal, the user inputs information related to their legal case. During this information input process, the emotion engine analyzes the user's facial expressions and voice to recognize their emotional state. The emotional state and legal case information are then sent to the server.

[1627] The server searches a database based on the received lawsuit information and sentiment data to extract similar past lawsuits. This search uses natural language processing (NLP) algorithms and similarity calculations. Using the analysis results of the extracted similar cases and sentiment data, a machine learning model (e.g., logistic regression or random forest) is used to calculate the winning rate.

[1628] Calculating winning rates and providing results

[1629] The server uses the emotion engine to take into account the user's emotional state and more accurately calculate the success rate of the lawsuit. A report is generated containing the calculation results and details of related similar cases. The report is displayed to the user through the user interface on their device, allowing them to view the success rate, similar cases, and specific advice. Based on the report and advice provided, the user decides whether to proceed with the lawsuit. Feedback from the emotion engine can also be used to help make legal decisions.

[1630] Generative AI advocacy

[1631] When legal amendments are made or new case data is added, the server collects court records and data on lawyers' legal activities and trains the generative AI. Based on the learned data, the generative AI acquires legal skills to support litigation.

[1632] When a user requests the Generative AI to defend their case, they re-enter the necessary legal information into their device and submit the request. During this process, the emotion engine recognizes the user's emotions and reflects that information in the Generative AI. The server then formulates an optimal defense strategy based on the received information and learning data. Once the lawsuit begins, the Generative AI will conduct defense activities in court and provide advice to the user in real time. The emotion engine continuously monitors the user's emotional state, and the Generative AI will adjust its strategy as needed.

[1633] Specific examples

[1634] For example, if a company reports a data breach incident, they can use the AI ​​security assistant application to input details of the incident. The emotion engine analyzes the stress and anxiety of the person in charge and compares it with similar cases in the database to conduct a risk assessment. The results are provided as a report, and any necessary mental support is also provided.

[1635] Example prompt sentence:

[1636] "Important data was leaked by an internal employee. As a countermeasure, we are considering strengthening log management and providing employee training."

[1637] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1638] Step 1:

[1639] The server collects past civil and fa...

Claims

1. A means of collecting and storing past case data on civil and family law cases in a database; a means for receiving the litigation information entered by the user and searching a database for similar past cases; A means for calculating the success rate of a lawsuit based on the searched similar cases and providing the result to the user; A system including:

2. A means of collecting court records and data on lawyers' legal activities and training the AI ​​to generate them; A means for a user to input information to request defense of the generated AI; The means by which generative AI can defend itself in court, The system of claim 1 , comprising:

3. A means for visually displaying the calculation results of the winning rate and details of similar cases through a user interface; a means for providing advice to users to help them decide whether to proceed with litigation; and The system of claim 1 , comprising:

Images