System

Abstract

A system is provided.SOLUTION: A system, comprising: means for a user to input a legal issue; means for managing a historical case database; means for searching for similar cases based on the legal issue; means for extracting certain features from search results; means for generating a plurality of predictive judgment patterns based on the extracted features; and means for presenting the generated predictive judgment patterns 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] In modern society, legal issues are becoming more complex and diverse, and the demand for legal advice is rapidly increasing. However, resolving legal issues typically requires consulting a lawyer, which is costly and time-consuming. This burden is particularly significant for individuals and small businesses. This invention aims to provide a means to quickly and cost-effectively provide predictions about legal issues and support users in making legal decisions. [Means for solving the problem]

[0005] In order to solve the above problems, the present invention provides a system including the following means: means for a user to input a legal problem, means for managing a database of past legal cases, means for searching for similar legal cases based on the legal problem, means for extracting specific features from the search results, means for generating a plurality of predicted decision patterns based on the extracted features, and means for presenting the generated predicted decision patterns to the user. Furthermore, by adding means for improving the accuracy of the generated predicted decision patterns using a machine learning algorithm and means for receiving user feedback and re-training the model, the reliability and utility of the system can be further improved.

[0006] "User" means any person or entity that uses the System to obtain a predicted decision on a legal matter.

[0007] "Input means" refers to the interface through which a user enters details of a legal matter into the system.

[0008] A "case law database" refers to data storage that accumulates and manages past court cases and related documents.

[0009] "Searching means" refers to an algorithm or program for searching a case law database based on a legal question entered by a user.

[0010] "Feature extraction means" refers to the processes and techniques used to extract important attributes and information from the retrieved case law data.

[0011] "Means for generating predicted verdict patterns" refers to algorithms and methods for predicting multiple possible verdicts based on extracted features.

[0012] "Presenting means" refers to an interface or method for displaying the generated predicted sentence pattern to a user.

[0013] A "machine learning algorithm" refers to a computational method that learns from past data and uses that knowledge to analyze and predict new data.

[0014] "Feedback" refers to the evaluations and opinions that users provide regarding the system's output.

[0015] "Retraining" refers to the process of improving the accuracy of a system's machine learning algorithms based on user feedback. [Brief explanation of the drawings]

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

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

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

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

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

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

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

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

[0024] [First embodiment]

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

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

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

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

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

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

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

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

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

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

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

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

[0037] The present invention is a system that utilizes a database of past legal precedents to provide predicted decisions on legal issues. The program processing of this system will be explained below in natural language.

[0038] 1. User Input Phase

[0039] The terminal provides the user with an interface for entering details of the legal issue, including the type of case, specific details of the issue (e.g., amount of compensation, breach of contract).

[0040] Through this interface, users enter details of the legal issue they wish to discuss, such as "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovations during the contract period."

[0041] 2. Data reception and retrieval by the server

[0042] The terminal transmits the data entered by the user to the server.

[0043] The server parses the received user-entered data and generates a search query. Parsing involves the process of extracting important keywords and phrases from the textual input data.

[0044] The server quickly searches the case law database based on the generated search query, and retrieves the top N related cases based on the similarity score and search criteria.

[0045] 3. Feature Extraction and Prediction

[0046] The server preprocesses the case data it receives, converting it into a format suitable for input to machine learning algorithms. This preprocessing includes text cleansing (removing unnecessary information and normalizing) and converting it into structured data.

[0047] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, the type of judgment, applicable laws, and sentencing.

[0048] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the feature values. This algorithm learns from past case data as training data and makes highly accurate predictions for input data.

[0049] 4. Presenting the prediction results

[0050] The server extracts and organizes the generated predicted judgment patterns and converts them into a format that is easy for the user to understand. Specifically, it presents specific judgment results such as "order to pay the full amount of unpaid rent," "order to pay repair costs and terminate the contract," and "payment of 300,000 yen as a settlement and continuation of the contract."

[0051] The terminal displays this predicted judgment result to the user, who can refer to it and use it as information for making legal decisions.

[0052] 5. Feedback and Retraining

[0053] The user can then provide feedback on the prediction results, which may include rating the accuracy of the prediction or providing additional information.

[0054] The server receives user feedback and retrains the machine learning model as needed, continuously improving the accuracy of the system.

[0055] As described above, the present invention provides users with an efficient and reliable tool for quickly assessing legal issues and obtaining a predicted verdict. Through specific implementation methods and processes, the present invention can provide assistance to individuals and businesses in effectively resolving legal issues.

[0056] The processing flow will be explained below.

[0057] Step 1: User Input

[0058] The terminal provides the user with an interface for entering details of the legal matter, including fields for the type of case, the issues at stake, and the specific circumstances.

[0059] Through the interface, the user inputs details of the legal issue they wish to discuss, such as breach of lease agreement, unpaid rent of 500,000 yen, and unauthorized renovations during the lease term.

[0060] The terminal transmits the data entered by the user to the server.

[0061] Step 2: Receiving and analyzing data

[0062] The server receives the user input data sent from the terminal.

[0063] The server analyzes the received data and extracts the type of data (breach of contract, claim for compensation, etc.) and related parameters (amount due, contract terms, etc.).

[0064] Step 3: Search the database

[0065] The server generates a search query based on the extracted data, which is used to search for relevant past cases in a case law database.

[0066] The server uses the search query to quickly search through the case database to retrieve cases similar to the user's legal problem, resulting in the top N most relevant cases.

[0067] Step 4: Feature extraction and preprocessing

[0068] The server analyzes the acquired case data and preprocesses it into a format suitable for machine learning algorithms, including text cleansing (removing unnecessary information and normalizing it) and feature extraction.

[0069] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, type of judgment, and sentence.

[0070] Step 5: Generate predicted verdict patterns

[0071] The server inputs the pre-processed and feature-extracted data into a machine learning algorithm. The machine learning model has already been trained based on past case data and generates multiple predictive verdict patterns for new data.

[0072] Predicted judgment patterns include, for example, "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and continuation of the contract."

[0073] Step 6: Presenting the results

[0074] The server organizes the generated predicted judgment patterns and converts them into a format that is easy for the user to understand.

[0075] The server transmits the predicted judgment result to the terminal.

[0076] The terminal displays the predicted verdict results sent from the server to the user, who can use this information to make legal decisions.

[0077] Step 7: Processing feedback

[0078] The user can then input feedback on the presented prediction results, which may include an evaluation of the prediction accuracy or additional detailed information.

[0079] The server receives user feedback and retrains the machine learning model as needed, continuously improving the system's prediction accuracy.

[0080] These are the specific processing steps of the present invention, which allow individuals and businesses to obtain predicted decisions on legal matters in an efficient and reliable manner.

[0081] Example 1

[0082] 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."

[0083] Currently, there is a lack of systems that provide fast and reliable predictive decisions on legal issues. Many legal issues are complex, and researching past cases and making appropriate legal decisions requires a lot of time and expertise. This makes it difficult for individuals and businesses to receive support to effectively resolve legal issues. In addition, existing systems often lack the accuracy of their machine learning algorithms or retraining capabilities, resulting in inaccurate predictions.

[0084] 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.

[0085] In this invention, the server includes: a means for a user to input legal issues; a means for managing a database of past legal precedents; a means for transmitting user-input data to the server; a means for analyzing the received user-input data and generating a search query; a means for searching the legal precedent database and retrieving relevant legal precedents based on the generated search query; a means for preprocessing the legal precedent data and extracting important features using natural language processing technology; a means for generating multiple predicted verdict patterns using a machine learning algorithm based on the extracted features; and a means for presenting the generated predicted verdict patterns to the user. This enables the server to provide quick and accurate predicted verdicts for legal issues input by the user. Furthermore, the accuracy of predictions can be continuously improved by retraining the machine learning model based on user feedback.

[0086] "User" means any person or entity that utilizes the System to enter legal matters and obtain predicted judgments.

[0087] "Legal Issue" means detailed information about the case or matter that a User seeks to resolve.

[0088] "Means for input" refers to an interface that allows a user to enter details of a legal issue in text form.

[0089] "Means for managing a case law database" refers to technologies and systems that store, search, and manage past case law information in a format that can be referenced.

[0090] "Means for transmitting data to a server" refers to the protocols and technologies used to transmit information entered by a user from a terminal to a server.

[0091] "Means for analyzing data" refers to technology used to process received user-entered data and extract required keywords and phrases.

[0092] "Means for generating a search query" refers to a technology for generating a query for searching a case law database based on the analyzed data.

[0093] The "means for searching a case law database" refers to a technology for quickly searching a case law database using a generated query to obtain relevant cases.

[0094] "Means for pre-processing data" refers to the process of preparing acquired case law data in a format suitable for input into machine learning algorithms.

[0095] "Means for extracting features" refers to technology that extracts important elements from case law data and organizes them into a format that can be used in machine learning models.

[0096] "Means for generating predicted verdict patterns" refers to the process of creating multiple predicted verdict patterns using a machine learning algorithm based on the extracted features.

[0097] The "means for presenting a predicted verdict pattern" refers to a technique for visually presenting the generated predicted verdict pattern to a user.

[0098] "Feedback" refers to the evaluation or additional information provided by the user in response to the predicted results.

[0099] "Retraining" refers to the process of retraining a machine learning model based on feedback from users.

[0100] MODE FOR CARRYING OUT THE INVENTION

[0101] The present invention is a system for providing predicted decisions on legal issues by utilizing a database of past legal cases. The system includes a means for a user to input legal issues, a means for managing the database of legal cases, a means for transmitting and receiving data, a means for analyzing data, a means for generating search queries, a means for searching legal case data, and a means for generating and presenting predicted decisions.

[0102] Hardware and Software Use

[0103] Terminal

[0104] The terminal provides an interface for users to enter legal information, for example, through a web browser or a dedicated application, and includes input fields where users can enter information using a keyboard or touchscreen.

[0105] server

[0106] The server receives the data entered by the user and parses it, using Python and NLP libraries (e.g., NLTK and spaCy) to parse the data and extract important keywords and phrases.

[0107] Based on the received data, a search query is generated for the case law database, and the server executes the generated query using a search engine such as Elasticsearch to retrieve relevant case law data.

[0108] The case data is preprocessed (using libraries such as Pandas and NumPy) and key features are extracted using natural language processing techniques (e.g., NLP models such as BERT).

[0109] Based on the extracted features, multiple predictive verdict patterns are generated using machine learning algorithms (using Scikit-learn and TensorFlow).

[0110] Prediction results

[0111] The server organizes the generated predicted verdict patterns and presents them in a user-friendly format. For example, the results are returned to the terminal in JSON format and rendered in HTML format on the terminal. The user can then view the results using a web browser or similar.

[0112] Feedback and model retraining

[0113] Users can provide feedback on the provided predictions, including a rating of the prediction's accuracy and additional information.

[0114] The server receives feedback and retrains the machine learning model as needed. The retraining process continuously improves the system's predictive accuracy.

[0115] Specific examples

[0116] For example, if a user writes "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," the server will analyze this information and search for similar past cases. The server extracts important features from the case data and applies a machine learning algorithm to present a predicted judgment, such as the following:

[0117] Order to pay all unpaid rent

[0118] "Order to pay repair costs and contract termination"

[0119] "Payment of 300,000 yen as settlement money and continuation of contract"

[0120] Example prompts for generative AI models

[0121] "Generate a predicted verdict based on past precedents for the following legal issue: a breach of lease agreement involving unpaid rent of 500,000 yen and unauthorized renovations during the lease term."

[0122] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0123] Step 1:

[0124] User Input Phase

[0125] The user uses the device interface to enter details of the legal issue, such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," into text input fields provided through a web browser or dedicated application.

[0126] Input: User-entered legal issue details text

[0127] Output: Structured data format (e.g. JSON) sent to the device

[0128] What happens: A web browser renders an HTML form, a user enters information using the keyboard, and clicks the submit button.

[0129] Step 2:

[0130] Data transmission

[0131] The device sends the legal details entered by the user to the server using the HTTPS protocol, encrypting the data.

[0132] Input: User-entered legal issue details

[0133] Output: Encrypted data packet sent to the server

[0134] Specific operation: The device generates an HTTP POST request and sends the data to the server, where the Web API receives the request.

[0135] Step 3:

[0136] Data Analysis and Query Generation

[0137] The server analyzes the received user-entered data, extracts key keywords and phrases, and generates a search query against the case law database.

[0138] Input: Legal issue details received from the device

[0139] Output: Search query against case law database

[0140] What it does: It uses Python and NLP libraries (e.g., NLTK or spaCy) on the server to extract important keywords and generate search queries based on those keywords.

[0141] Step 4:

[0142] Case Law Database Search

[0143] The server searches the case law database based on the generated search query to retrieve relevant cases, calculates a similarity score, and selects the most relevant case.

[0144] Input: search query

[0145] Output: Relevant case law data

[0146] What happens: The server uses a search engine such as Elasticsearch to issue a query and search the database.

[0147] Step 5:

[0148] Data Preprocessing

[0149] The server preprocesses the case data it receives, converting it into a format suitable for input to machine learning algorithms. Preprocessing includes text cleansing and normalization.

[0150] Input: Retrieved case data

[0151] Output: Preprocessed data format

[0152] Specific operation: Using libraries such as Pandas and NumPy on the server side, unnecessary information is removed and the data is normalized.

[0153] Step 6:

[0154] Feature extraction

[0155] The server uses natural language processing techniques to extract important features from the case data.

[0156] Input: Preprocessed case data

[0157] Output: Feature data

[0158] How it works: An NLP model (e.g., BERT) on the server analyzes the text data and extracts important features such as the issues at stake and the type of judgment.

[0159] Step 7:

[0160] Predictive Verdict Generation

[0161] The server generates multiple predicted verdict patterns using a machine learning algorithm based on the extracted features.

[0162] Input: Feature data

[0163] Output: Predicted verdict pattern

[0164] Specific operation: The server runs machine learning models using Scikit-learn and TensorFlow to generate predicted verdict patterns.

[0165] Step 8:

[0166] Prediction results

[0167] The server organizes the generated predicted verdict patterns and sends them to the terminal for presentation to the user. The results are displayed in a format that is easy for the user to understand.

[0168] Input: Predicted sentence pattern

[0169] Output: The resulting data that is presented to the user.

[0170] Specific operation: The result data is sent to the terminal in JSON format, converted to HTML format on the terminal, and displayed. The user can check the results using a web browser or similar.

[0171] Step 9:

[0172] Gathering feedback

[0173] The user can then provide feedback on the provided predictions, including a rating on the accuracy of the predictions and additional information.

[0174] Input: User feedback data

[0175] Output: Feedback data sent to the server

[0176] Specific behavior: The user enters feedback information using a web form on the device and clicks the submit button.

[0177] Step 10:

[0178] Retraining the Model

[0179] The server receives feedback from users and retrains the machine learning model as needed, improving prediction accuracy.

[0180] Input: Feedback data

[0181] Output: Retrained machine learning model

[0182] How it works: The server inputs the feedback data into the machine learning algorithm and retrains the model. The retraining process is performed periodically as a batch process.

[0183] (Application example 1)

[0184] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0185] Systems that provide predicted decisions for legal issues provide important information by utilizing databases of past legal precedents, but they have difficulty accurately assessing legal risks related to security risks. Another challenge is continually improving the accuracy of predicted decisions based on information entered by users. Furthermore, it is necessary to continuously improve accuracy by incorporating re-learning that reflects user feedback into the system. There is a need to solve these issues.

[0186] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0187] In this invention, the server includes means for a user to input a legal problem, means for managing a database of past legal cases, means for searching for similar legal cases based on the legal problem, means for extracting specific features from the search results, means for generating a plurality of predicted verdict patterns based on the extracted features, means for presenting the generated predicted verdict patterns to the user, means for inputting details of security risks, and means for performing legal risk assessment, thereby enabling users to predict verdicts for legal problems and security risks with high accuracy.

[0188] Definitions of important terms contained in the claims

[0189] "User" means any person or entity that utilizes the System to input legal issues and security risks and receive predicted adjudications and legal risk assessments.

[0190] A "legal issue" is a legal issue, dispute, or controversy that seeks a legal solution to a specific matter.

[0191] A "case law database" is a database that collects and organizes data on past trials, legal decisions, and verdicts, and is used to resolve similar legal issues.

[0192] "Similar cases" are past cases that have content very similar to the legal issue entered by the user.

[0193] "Features" refer to important elements and parameters extracted from the case law database, and serve as the basic data for predicting verdicts.

[0194] A "predicted judgment pattern" is a set of multiple judgment outcomes generated by a machine learning algorithm based on past case data for a given legal issue.

[0195] "Security risks" refer to threats and dangers to the security of companies and individuals, such as information leaks, system hacking, and internal crimes.

[0196] "Legal risk assessment" is the process of predicting and evaluating the legal consequences and impacts of security risks.

[0197] "Feedback" refers to opinions and suggestions for improvement regarding the use of the system provided by users, which are reflected in improving the accuracy of the system.

[0198] "Retraining" is the process of updating an existing machine learning model to improve its accuracy and performance based on user feedback and additional information.

[0199] MODE FOR CARRYING OUT THE INVENTION

[0200] In carrying out the present invention, a system for assessing legal issues and security risks is constructed, and a specific embodiment thereof is shown below.

[0201] First, the user uses a terminal to input details of legal issues and security risks. This input is done through a dedicated interface. For example, for legal issues, users can enter information such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovations during the contract period," while for security risks, users can enter details such as "Information leak," "System hacking," and "Employee internal crime."

[0202] The entered data is received and analyzed by the server. The server extracts important keywords and phrases from the text data and generates a search query based on these. The server manages a database of past legal precedents and uses the generated search query to quickly search for relevant legal precedents.

[0203] Next, the server extracts specific features from the search results. Natural language processing (NLP) technology is used to extract these features. Preprocessing is then performed, including text cleansing (removing unnecessary information and normalizing it) and conversion to structured data. Then, a machine learning algorithm is used to analyze the data based on the features, and multiple predictive verdict patterns are generated.

[0204] The generated predicted judgment patterns are presented to the user. Specific judgment outcomes, such as "order to pay all outstanding rent," "order to pay repair costs and terminate the contract," or "pay 300,000 yen as a settlement and continue the contract," are displayed as prediction results. Furthermore, a legal risk assessment of security risks is also presented, enabling companies and individuals to take appropriate countermeasures.

[0205] Users can input feedback on these prediction results on their devices, and the server retrains the machine learning model based on the received feedback. This retraining process continuously improves the accuracy of the system.

[0206] The specific implementation of this system uses the following hardware and software: a cloud server (e.g., AWS EC2, Google Cloud), a web framework (Flask), a machine learning library (Sklearn), and a library for saving and loading models (joblib).

[0207] For example, if a user inputs "a past case of a breach of a rental contract resulting in unpaid rent of 500,000 yen," the server will search the database of past legal precedents to extract similar cases. It will then present predicted judgment patterns such as "an order to pay the full amount of unpaid rent" or "an order to pay repair costs and terminate the contract," allowing the user to quickly obtain a legal decision.

[0208] Example prompt sentence:

[0209] "Unpaid rent of 500,000 yen, unauthorized renovations during the rental contract"

[0210] In this manner, the system of the present invention provides users with predicted verdicts and assessments of legal issues and security risks, providing legal decision support.

[0211] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0212] Processing steps of the system that realizes the application example

[0213] Step 1:

[0214] The user enters details about the legal issue or security risk.

[0215] The user enters the specific type of case and details of the problem (e.g., "Violation of rental contract" or "Unpaid rent: 500,000 yen") into the input field on the terminal.

[0216] Step 2:

[0217] The terminal transmits the input data to the server.

[0218] The terminal captures data through the input interface and transmits the data to the server using a secure communication protocol (such as HTTPS).

[0219] Step 3:

[0220] The server analyzes the received data and generates a search query.

[0221] The server uses natural language processing (NLP) techniques to extract important keywords and phrases from the text data, which then generate the search query.

[0222] Step 4:

[0223] The server searches a database of past legal precedents using the generated search query.

[0224] The server uses a high-speed search engine to search the database to identify similar cases, and retrieves the top N relevant cases based on the search criteria.

[0225] Step 5:

[0226] The server extracts specific features from the acquired case data.

[0227] The server preprocesses the data into a format suitable for machine learning algorithms, including text normalization and removal of unnecessary information, and extracts key features (e.g., issues at issue, type of decision, applicable law, sentencing, etc.).

[0228] Step 6:

[0229] The server generates a predicted verdict pattern based on the extracted features.

[0230] The server uses a machine learning model to generate multiple predictive judgment patterns using the extracted features as input. The model is a generative AI model that has learned from past case data.

[0231] Step 7:

[0232] The server presents the generated predicted sentence pattern to the user.

[0233] The server converts the prediction results into a user-friendly format and sends them to the terminal, where the user can check the predicted verdict results on the terminal interface.

[0234] Step 8:

[0235] The user inputs feedback on the provided prediction results.

[0236] The user inputs feedback about the accuracy of the prediction results and additional information into the terminal.

[0237] Step 9:

[0238] The server receives user feedback and retrains the system.

[0239] The server analyzes the feedback data and creates a retraining dataset, which improves the accuracy of the machine learning model.

[0240] In this way, the system provides users with predictive adjudications on legal issues and security risks, and continually improves its accuracy.

[0241] 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.

[0242] The present invention is a system that provides predicted decisions on legal issues by combining a database of past legal precedents with an emotion engine that identifies user emotions. The program processing of this system is explained below in natural language.

[0243] 1. User Input Phase

[0244] The device provides a user with an interface for inputting details of the legal issue and emotional state. The input fields include the type of case, specific details of the issue (e.g., amount of compensation, breach of contract), etc. It also includes sensors and question-type items for inputting or detecting the user's emotional state.

[0245] Through this interface, users can enter details of the legal issue they wish to discuss and, if necessary, their emotional state. For example, they can enter information such as "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," as well as their own emotional state (e.g., anger, anxiety).

[0246] The terminal transmits the data entered by the user and the detected emotional state to the server.

[0247] 2. Data reception and analysis by the server

[0248] The server receives the user's input data and emotion data transmitted from the terminal.

[0249] The server analyzes the received data and extracts the type of data (e.g., breach of contract, claim for compensation, etc.) and related parameters (e.g., outstanding amount, contract terms, etc.). It also analyzes the user's emotional state using an emotion engine.

[0250] 3. Database Search

[0251] The server generates a search query based on the analysis results, which is used to search for relevant past cases in a case law database.

[0252] The server uses the search query to rapidly search the case database to retrieve cases similar to the user's legal problem, resulting in the top N most relevant cases.

[0253] 4. Feature Extraction and Preprocessing

[0254] The server preprocesses the case data and converts it into a format suitable for machine learning algorithms. Preprocessing includes text cleansing (removing unnecessary information and normalizing) and feature extraction.

[0255] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, type of judgment, and sentence.

[0256] 5. Generating predicted verdict patterns

[0257] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the pre-processed and feature-extracted data and the user's emotional state. The machine learning model has already been trained based on past case data and can make highly accurate predictions for new data.

[0258] Predicted judgment patterns include, for example, "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and continuation of the contract."

[0259] 6. Presentation of results

[0260] The server selects the predicted judgment pattern that best suits the user's emotional state based on the emotion engine, thereby providing information in a form that is most acceptable to the user.

[0261] The server transmits the optimal predicted judgment pattern to the terminal.

[0262] The terminal displays the predicted verdict results sent from the server to the user, who can use this information to make legal decisions.

[0263] 7. Feedback and Retraining

[0264] The user can then provide feedback on the provided predictions, which may include a rating on the accuracy of the predictions or additional details.

[0265] The server receives user feedback and retrains the machine learning model as needed, continuously improving the system's prediction accuracy. This retraining also incorporates feedback based on the user's emotional state.

[0266] As a result, the present invention not only provides a predictive verdict for a user's legal problem, but also takes into account the user's emotional state to provide more appropriate advice in a fast and reliable manner. Specific implementation methods and processes are provided to help individuals and businesses effectively resolve legal issues.

[0267] The processing flow will be explained below.

[0268] Step 1: User Input

[0269] The terminal provides a user with an interface for entering details of the legal issue and emotional state, including fields for the type of case, details of the issue (e.g., amount of compensation, breach of contract).

[0270] Through the interface, users enter details of the legal issue they wish to discuss, and also enter their emotional state (e.g., anger, anxiety, etc.) in a field reflecting the emotional state.

[0271] The device sends the details of the legal issue and the emotional state entered by the user to the server in one batch.

[0272] Step 2: Receiving and analyzing data

[0273] The server receives the user's input data and emotion data transmitted from the terminal.

[0274] The server analyzes the received data and extracts the type of input data (e.g., breach of contract, claim for compensation, etc.) and related parameters (e.g., unpaid amount, contract terms, etc.). It also uses an emotion engine to analyze the user's emotional state.

[0275] Step 3: Search the database

[0276] The server generates a search query based on the analysis results, which includes specific keywords related to the law and emotional state.

[0277] The server uses the search query to quickly search the case law database, extracting cases similar to the user's legal problem, and retrieves the top N most relevant cases as search results.

[0278] Step 4: Feature extraction and preprocessing

[0279] The server analyzes the acquired case data and performs preprocessing, which includes text cleansing (removing unnecessary information and normalizing) and feature extraction.

[0280] The server uses natural language processing (NLP) technology to extract important features from case data, including the issues at issue, the type of judgment, and the sentence.

[0281] Step 5: Generate predicted verdict patterns

[0282] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the results of preprocessing and feature extraction and the user's emotional state. The algorithm is trained based on past case data and can make highly accurate predictions for new data.

[0283] Predicted judgment patterns are generated in specific forms, such as "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and the contract continue."

[0284] Step 6: Presenting the results

[0285] The server selects a predictive judgment pattern that is appropriate for the user's emotional state based on the emotion engine, and this selection provides information in a form that is most acceptable to the user.

[0286] The server transmits the optimal predicted judgment pattern to the terminal.

[0287] The terminal displays the predicted judgment results sent from the server to the user, who then makes legal decisions based on the displayed information.

[0288] Step 7: Processing feedback

[0289] The user can then input feedback on the presented prediction results, which may include an evaluation of the prediction accuracy or additional detailed information.

[0290] The server receives user feedback and retrains the machine learning model as needed, including feedback based on the user's emotional state, to continuously improve the system's prediction accuracy.

[0291] Through these specific processing steps, the present invention can provide a predicted judgment for the user's legal problem and further provide an optimal solution by taking the user's feelings into consideration.

[0292] Example 2

[0293] 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."

[0294] Conventional systems for legal issues are unable to provide predicted decisions that take into account the user's emotional state, making it difficult to provide optimal legal advice quickly. Furthermore, if the accuracy of the predicted decisions provided is low, there is no mechanism to improve the system's performance by incorporating user feedback, which limits the system's usefulness.

[0295] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for a user to input details of a legal issue and their emotional state, means for analyzing received data based on the legal issue and their emotional state, means for managing a database of past legal cases and searching for similar legal cases, means for extracting specific features from the search results and performing preprocessing, means for using a machine learning algorithm to generate multiple predicted verdict patterns based on the extracted features and the user's emotional state, means for selecting and presenting the generated predicted verdict patterns in an optimal form based on the user's emotional state, and means for collecting user feedback and performing re-learning to continuously improve the accuracy of the system. This makes it possible to quickly provide highly accurate predicted verdicts while taking the user's emotional state into consideration.

[0296] "Legal issue details" refers to the specific legal matter the user is attempting to address, including the type of matter and specific details of the issue (e.g., amount of compensation, details of contract breach).

[0297] "Emotional state" refers to the user's current mental or emotional state, and is information that expresses emotions such as "anger," "anxiety," and "sadness."

[0298] "Means for analysis" refers to the technical means for carrying out the process of analyzing input data and extracting data types and important parameters.

[0299] A "case law database" is a database system that systematically accumulates past legal decisions and court cases and manages them in a searchable format.

[0300] A "searching means" is a technical means for efficiently locating information in a database based on specific criteria.

[0301] "Features" are important attributes or metrics extracted from a dataset that machine learning models use to make predictions or classifications.

[0302] "Preprocessing" refers to processes such as data cleansing and normalization that transform data into a form suitable for analytics and machine learning algorithms.

[0303] A "machine learning algorithm" is a mathematical and statistical method for learning from data and making predictions or classifications.

[0304] A "predicted judgment pattern" is a pattern that indicates possible future court outcomes, generated based on analyzed data.

[0305] "Feedback" refers to user-provided evaluations and additional information about the system's predictive accuracy and usability.

[0306] "Retraining" is the process of updating a machine learning model and improving its performance based on new data and feedback collected.

[0307] The present invention is a system that provides predicted decisions for users' legal issues. The system combines a database of past legal cases with an emotion engine that identifies user emotions to provide more reliable legal advice.

[0308] The main components of the system are:

[0309] 1. User Interface - The terminal provides the user with an interface for entering details of the legal issue and emotional state. This interface includes fields for entering the case type, specific details of the issue, and emotional state.

[0310] 2. Data Reception and Analysis - The server receives the user's input data and emotion data sent from the device. The received data is analyzed using analysis software to extract the data type and important parameters. The emotion engine is also used to analyze the user's emotional state.

[0311] 3. Case database search - The server generates a search query based on the received data and analysis results, and searches the case database for the most relevant top N cases.

[0312] 4. Preprocessing and Feature Extraction - The server preprocesses the extracted case data, performing text cleansing and normalization, and then uses natural language processing (NLP) techniques to extract features such as the issues at issue, type of verdict, and sentence.

[0313] 5. Generation of predicted verdict patterns - The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the pre-processed data and the user's emotional state. This machine learning algorithm is trained based on past case data, allowing for highly accurate predictions.

[0314] 6. Presentation of results - The server selects the predicted verdict pattern that best suits the user's emotional state and sends it to the terminal. The terminal displays the received predicted verdict results to the user, allowing the user to make legal decisions based on this information.

[0315] 7. Feedback collection and retraining - The user inputs feedback on the provided predicted verdict results into the terminal. The server receives this feedback and retrains the machine learning model to improve the system's prediction accuracy.

[0316] Specific examples

[0317] For example, suppose a user wants to review a "violation of a rental contract." The user enters the following information into their terminal: "Unpaid rent: 500,000 yen," "Unauthorized renovation during the contract period," and selects "anger" as their emotional state. This information is sent to the server, where it is analyzed and searched through a precedent database. A predicted judgment pattern is generated based on the user's input and emotional state and presented to the user in the form of, for example, "Order to pay the full amount of unpaid rent," "Order to pay repair costs and terminate the contract," or "Pay 300,000 yen as a settlement and continue the contract."

[0318] Prompt Sentence Examples

[0319] An example of a prompt to input to a generative AI model is as follows:

[0320] text

[0321] User Input: Rental Agreement Violation

[0322] Problem details: Unpaid rent: 500,000 yen, unauthorized renovations during the contract period

[0323] Emotional state: Anger

[0324] System output: Judgment pattern 1: Order to pay the full amount of unpaid rent. Judgment pattern 2: Order to pay repair costs and terminate the contract. Judgment pattern 3: Payment of 300,000 yen as a settlement and continuation of the contract.

[0325] This prompt shows how the system will predict the outcome based on the information entered by the user. As can be seen from this example, the system takes into account the user's emotional state to provide the optimal predicted verdict, enabling it to provide fast and reliable legal advice.

[0326] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0327] Step 1: User Input Phase

[0328] The user inputs details of the legal issue and their emotional state into the terminal. Specifically, they input information such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period" through the user interface. They also select "Anger" as their emotional state. The input fields are equipped with text boxes and drop-down menus. This input data is temporarily stored in the terminal's memory and prepared for the next processing step.

[0329] Input: Details of the legal issue and emotional state entered by the user into the device

[0330] Output: User input data sent from the device to the server

[0331] Step 2: Data reception and analysis

[0332] The server receives user data sent from the device. This data is sent to the server in a structured data format such as JSON. The server uses data analysis software to analyze the input data and extract the type of legal issue and important parameters. It also uses an emotion engine to analyze the user's emotional state.

[0333] Input: User input data sent from the device to the server

[0334] Output: Analyzed legal issue type, parameters, and emotional state

[0335] Step 3: Search the database

[0336] The server generates a search query based on the analysis results. The generated query is used to search the case law database. For example, conditions such as "Violation of rental contract" and "Unpaid rent: 500,000 yen" are used as a query. The server sends this query to a database management system (DBMS) and quickly searches for related past cases.

[0337] Input: A search query generated based on the analysis results

[0338] Output: A set of relevant case law data

[0339] Step 4: Feature extraction and preprocessing

[0340] The server preprocesses the case data it receives. This preprocessing includes text cleansing and normalization. It uses natural language processing (NLP) techniques to extract features from the case data, such as the issues at issue, type of judgment, and sentence. This converts the data into a format suitable for machine learning algorithms.

[0341] Input: Searched case data

[0342] Output: Preprocessed and feature extracted data

[0343] Step 5: Generate predicted sentence patterns

[0344] The server uses a machine learning algorithm to generate multiple predictive verdict patterns based on the preprocessed and feature-extracted data and the user's emotional state. Because the machine learning model has already been trained on past case data, it can make highly accurate predictions for new data.

[0345] Input: Preprocessed and feature-extracted data, user emotional state

[0346] Output: A list of predicted verdict patterns

[0347] Step 6: Presenting the results

[0348] The server selects the predicted verdict pattern that best suits the user's emotional state from the generated patterns and sends it to the terminal. The terminal then displays the received predicted verdict results to the user, who can then make legal decisions based on this information.

[0349] Input: List of predicted sentence patterns

[0350] Output: The optimal predicted verdict pattern presented to the user.

[0351] Step 7: Feedback and Retraining

[0352] The user inputs feedback on the provided predicted verdict results into the terminal. The feedback includes an evaluation of the prediction accuracy and additional detailed information. The server receives this feedback and retrains the machine learning model. This retraining process continuously improves the system's prediction accuracy.

[0353] Input: User feedback

[0354] Output: The machine learning model after retraining

[0355] Through the above processing steps, the system can quickly provide highly accurate predictive verdicts that take into account the user's emotional state.

[0356] (Application example 2)

[0357] 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."

[0358] Conventional legal problem-solving systems only refer to past legal precedents for legal problems entered by users, and do not provide predicted verdicts that take the user's emotional state into account. This makes it difficult to provide predicted verdicts that are intuitive and easy for users to accept. Furthermore, while security operations require rapid and appropriate responses, conventional systems lack the functionality to support this.

[0359] 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.

[0360] In this invention, the server includes means for a user to input a legal issue and emotional state, means for managing a database of past legal cases, means for searching for similar legal cases based on the legal issue and emotional state, means for extracting specific features from the search results, means for generating a plurality of predicted verdict patterns based on the extracted features and the emotional state, means for presenting the generated predicted verdict patterns to the user, and means for selecting a predicted verdict pattern that best suits the user's emotional state, thereby making it possible to provide an effective and appropriate predicted verdict that takes into account the user's emotional state.

[0361] A "means for user input of legal issues and emotional state" is a device or software that provides an interface through which a user can input or detect details of a specific legal issue and their emotional state.

[0362] The "means for managing a database of past court cases" is a database system for storing past court case data and for efficiently searching and managing it.

[0363] The "means for searching for similar precedents based on the legal issue and emotional state" refers to an algorithm or technology that searches for relevant past precedents based on the legal issue and emotional state entered by the user.

[0364] "Means for extracting specific features from search results" refers to natural language processing and data analysis techniques for extracting important parameters and characteristics from the searched case law data.

[0365] The "means for generating a plurality of predicted verdict patterns based on the extracted features and the emotional state" refers to a machine learning model or algorithm that generates a plurality of possible predicted verdicts using the extracted data and the user's emotional state.

[0366] The "means for presenting the generated predicted judgment pattern to the user" refers to a screen or notification system that displays the predicted judgment result in an easy-to-understand manner for the user.

[0367] The "means for selecting a predicted verdict pattern that is most suitable for the user's emotional state" refers to an algorithm or system that takes into account the user's current emotional state and selects it to present the most appropriate predicted verdict.

[0368] "Means for improving the accuracy of generated predicted verdict patterns using a generative AI model" refers to technologies and methods that use a pre-trained generative AI model to improve the accuracy of the predicted verdicts provided.

[0369] The "means for receiving the user's feedback and re-training the generative AI model" refers to a method for improving prediction accuracy by collecting feedback provided by users and re-training the generative AI model based on that feedback.

[0370] This invention is a security operations support system that combines an interface for inputting users' legal issues and emotional state with a system for managing, searching, and analyzing past case law data. This system can be installed on smartphones, smart glasses, head-mounted displays, and robots. A specific embodiment of this system is described below.

[0371] Hardware and Software

[0372] Smartphone: Provides an interface for users to enter details of their legal issue and their emotional state.

[0373] Server: Receives and analyzes data, searches the case law database, extracts features, generates predicted judgment patterns, and presents the results to the user.

[0374] Natural Language Processing (NLP) technology: Used to extract specific features from retrieved cases.

[0375] Machine learning models: Used to generate multiple predictive verdict patterns based on extracted features and emotional states, including logistic regression and generative AI models.

[0376] TextBlob: A library used to parse emotional states from user input.

[0377] Database management system: Used to efficiently manage past case data.

[0378] Processing Step Description

[0379] 1. User Input Phase: The user inputs details of the specific legal problem and their emotional state through a smartphone interface. For example, they input the text, "I spotted a suspicious person while on night patrol. I was attacked by the person and temporarily hid. The situation was very dangerous, so I called the police." and their emotional state, "I was very anxious and confused."

[0380] 2. Data reception and analysis by the server: The server receives the data entered by the user and extracts features using natural language processing techniques (such as TfidfVectorizer). At the same time, it analyzes the emotional state using TextBlob.

[0381] 3. Database Search: The server searches similar past cases based on the legal issue and emotional state, thereby obtaining relevant cases that match the issue the user is facing.

[0382] 4. Feature extraction and preprocessing: Extract important features (e.g., issues at issue, type of judgment, sentence, etc.) from the retrieved cases and convert them into a format suitable for machine learning models (logistic regression, generative AI models, etc.).

[0383] 5. Generation of predicted judgment patterns: Based on the extracted features and emotional state, a machine learning model is used to generate multiple predicted judgment patterns, such as "Order to pay repair costs and terminate the contract," "Order to pay the full amount of unpaid rent," and "Pay 300,000 yen as a settlement and continue the contract."

[0384] 6. Presentation of results: The server selects the predicted judgment pattern that best suits the user's emotional state from the generated patterns and presents it to the user through the smartphone interface.

[0385] Specific prompt examples

[0386] As a concrete example, the system operates based on the following prompt sentence:

[0387] Case details: "While on night patrol, I spotted a suspicious individual. I was attacked by the individual and temporarily hid. The situation was extremely dangerous, so I called the police."

[0388] Emotional state: "Very anxious and confused."

[0389] Based on this prompt, the system generates an appropriate predictive verdict and provides users with appropriate and specific countermeasures, enabling quick and effective responses in the field of security operations.

[0390] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0391] Step 1: User Input Phase

[0392] Subject: User

[0393] Specific Actions: Users input details of specific legal issues and their emotional state through a smartphone interface.

[0394] Input: For example, the text "I spotted a suspicious person while on night patrol. I was attacked by the person and temporarily hid. The situation was very dangerous, so I called the police." and the emotional state "I was very anxious and confused."

[0395] Output: User input data (legal issue details and emotional state).

[0396] Step 2: Send data

[0397] Subject: Device

[0398] Specific operation: The terminal sends the data entered by the user to the server.

[0399] Input: Details of the legal issue and emotional state entered by the user.

[0400] Output: The user's input data received by the server.

[0401] Step 3: Data reception and analysis

[0402] Subject: Server

[0403] Specific operation: The server receives the user's input data sent from the device, analyzes the received data, extracts features using natural language processing techniques (such as TfidfVectorizer), and analyzes the emotional state using TextBlob.

[0404] Input: User input data sent by the device.

[0405] Output: Extracted features and parsed emotional states.

[0406] Step 4: Search the case law database

[0407] Subject: Server

[0408] Specific operation: The server searches for similar past cases based on the legal issue and emotional state, thereby obtaining relevant cases that match the issue the user is facing.

[0409] Input: Extracted features and parsed emotional states.

[0410] Output: Retrieved relevant case law data.

[0411] Step 5: Feature extraction and preprocessing

[0412] Subject: Server

[0413] Specific operation: The server extracts important features (e.g., issues at issue, type of judgment, sentence, etc.) from the retrieved case data and converts them into a format suitable for machine learning models (logistic regression, generative AI models, etc.).

[0414] Input: The relevant case law data retrieved.

[0415] Output: Preprocessed feature data.

[0416] Step 6: Generate predicted sentence patterns

[0417] Subject: Server

[0418] Specific operation: Based on the extracted features and emotional state, the server uses a machine learning model to generate multiple predicted judgment patterns, such as "order to pay repair costs and terminate the contract," "order to pay the full amount of unpaid rent," and "pay 300,000 yen as a settlement and continue the contract."

[0419] Input: Preprocessed feature data and parsed emotional states.

[0420] Output: Multiple predicted verdict patterns.

[0421] Step 7: Select and present results

[0422] Subject: Server

[0423] Specific operation: The server selects the predicted judgment pattern that best suits the user's emotional state from the generated predicted judgment patterns and presents it to the user through the smartphone interface.

[0424] Input: Multiple predicted judgment patterns and analyzed emotional states.

[0425] Output: The predicted verdict pattern displayed to the user.

[0426] 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.

[0427] 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.

[0428] 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.

[0429] [Second embodiment]

[0430] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0431] 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.

[0432] 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).

[0433] 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.

[0434] 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.

[0435] 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).

[0436] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0437] 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.

[0438] 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.

[0439] 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.

[0440] 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.

[0441] 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."

[0442] The present invention is a system that utilizes a database of past legal precedents to provide predicted decisions on legal issues. The program processing of this system will be explained below in natural language.

[0443] 1. User Input Phase

[0444] The terminal provides the user with an interface for entering details of the legal issue, including the type of case, specific details of the issue (e.g., amount of compensation, breach of contract).

[0445] Through this interface, users enter details of the legal issue they wish to discuss, such as "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovations during the contract period."

[0446] 2. Data reception and retrieval by the server

[0447] The terminal transmits the data entered by the user to the server.

[0448] The server parses the received user-entered data and generates a search query. Parsing involves the process of extracting important keywords and phrases from the textual input data.

[0449] The server quickly searches the case law database based on the generated search query, and retrieves the top N related cases based on the similarity score and search criteria.

[0450] 3. Feature Extraction and Prediction

[0451] The server preprocesses the case data it receives, converting it into a format suitable for input to machine learning algorithms. This preprocessing includes text cleansing (removing unnecessary information and normalizing) and converting it into structured data.

[0452] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, the type of judgment, applicable laws, and sentencing.

[0453] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the feature values. This algorithm learns from past case data as training data and makes highly accurate predictions for input data.

[0454] 4. Presenting the prediction results

[0455] The server extracts and organizes the generated predicted judgment patterns and converts them into a format that is easy for the user to understand. Specifically, it presents specific judgment results such as "order to pay the full amount of unpaid rent," "order to pay repair costs and terminate the contract," and "payment of 300,000 yen as a settlement and continuation of the contract."

[0456] The terminal displays this predicted judgment result to the user, who can refer to it and use it as information for making legal decisions.

[0457] 5. Feedback and Retraining

[0458] The user can then provide feedback on the prediction results, which may include rating the accuracy of the prediction or providing additional information.

[0459] The server receives user feedback and retrains the machine learning model as needed, continuously improving the accuracy of the system.

[0460] As described above, the present invention provides users with an efficient and reliable tool for quickly assessing legal issues and obtaining a predicted verdict. Through specific implementation methods and processes, the present invention can provide assistance to individuals and businesses in effectively resolving legal issues.

[0461] The processing flow will be explained below.

[0462] Step 1: User Input

[0463] The terminal provides the user with an interface for entering details of the legal matter, including fields for the type of case, the issues at stake, and the specific circumstances.

[0464] Through the interface, the user inputs details of the legal issue they wish to discuss, such as breach of lease agreement, unpaid rent of 500,000 yen, and unauthorized renovations during the lease term.

[0465] The terminal transmits the data entered by the user to the server.

[0466] Step 2: Receiving and analyzing data

[0467] The server receives the user input data sent from the terminal.

[0468] The server analyzes the received data and extracts the type of data (breach of contract, claim for compensation, etc.) and related parameters (amount due, contract terms, etc.).

[0469] Step 3: Search the database

[0470] The server generates a search query based on the extracted data, which is used to search for relevant past cases in a case law database.

[0471] The server uses the search query to quickly search through the case database to retrieve cases similar to the user's legal problem, resulting in the top N most relevant cases.

[0472] Step 4: Feature extraction and preprocessing

[0473] The server analyzes the acquired case data and preprocesses it into a format suitable for machine learning algorithms, including text cleansing (removing unnecessary information and normalizing it) and feature extraction.

[0474] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, type of judgment, and sentence.

[0475] Step 5: Generate predicted verdict patterns

[0476] The server inputs the pre-processed and feature-extracted data into a machine learning algorithm. The machine learning model has already been trained based on past case data and generates multiple predictive verdict patterns for new data.

[0477] Predicted judgment patterns include, for example, "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and continuation of the contract."

[0478] Step 6: Presenting the results

[0479] The server organizes the generated predicted judgment patterns and converts them into a format that is easy for the user to understand.

[0480] The server transmits the predicted judgment result to the terminal.

[0481] The terminal displays the predicted verdict results sent from the server to the user, who can use this information to make legal decisions.

[0482] Step 7: Processing feedback

[0483] The user can then input feedback on the presented prediction results, which may include an evaluation of the prediction accuracy or additional detailed information.

[0484] The server receives user feedback and retrains the machine learning model as needed, continuously improving the system's prediction accuracy.

[0485] These are the specific processing steps of the present invention, which allow individuals and businesses to obtain predicted decisions on legal matters in an efficient and reliable manner.

[0486] Example 1

[0487] 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."

[0488] Currently, there is a lack of systems that provide fast and reliable predictive decisions on legal issues. Many legal issues are complex, and researching past cases and making appropriate legal decisions requires a lot of time and expertise. This makes it difficult for individuals and businesses to receive support to effectively resolve legal issues. In addition, existing systems often lack the accuracy of their machine learning algorithms or retraining capabilities, resulting in inaccurate predictions.

[0489] 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.

[0490] In this invention, the server includes: a means for a user to input legal issues; a means for managing a database of past legal precedents; a means for transmitting user-input data to the server; a means for analyzing the received user-input data and generating a search query; a means for searching the legal precedent database and retrieving relevant legal precedents based on the generated search query; a means for preprocessing the legal precedent data and extracting important features using natural language processing technology; a means for generating multiple predicted verdict patterns using a machine learning algorithm based on the extracted features; and a means for presenting the generated predicted verdict patterns to the user. This enables the server to provide quick and accurate predicted verdicts for legal issues input by the user. Furthermore, the accuracy of predictions can be continuously improved by retraining the machine learning model based on user feedback.

[0491] "User" means any person or entity that utilizes the System to enter legal matters and obtain predicted judgments.

[0492] "Legal Issue" means detailed information about the case or matter that a User seeks to resolve.

[0493] "Means for input" refers to an interface that allows a user to enter details of a legal issue in text form.

[0494] "Means for managing a case law database" refers to technologies and systems that store, search, and manage past case law information in a format that can be referenced.

[0495] "Means for transmitting data to a server" refers to the protocols and technologies used to transmit information entered by a user from a terminal to a server.

[0496] "Means for analyzing data" refers to technology used to process received user-entered data and extract required keywords and phrases.

[0497] "Means for generating a search query" refers to a technology for generating a query for searching a case law database based on the analyzed data.

[0498] The "means for searching a case law database" refers to a technology for quickly searching a case law database using a generated query to obtain relevant cases.

[0499] "Means for pre-processing data" refers to the process of preparing acquired case law data in a format suitable for input into machine learning algorithms.

[0500] "Means for extracting features" refers to technology that extracts important elements from case law data and organizes them into a format that can be used in machine learning models.

[0501] "Means for generating predicted verdict patterns" refers to the process of creating multiple predicted verdict patterns using a machine learning algorithm based on the extracted features.

[0502] The "means for presenting a predicted verdict pattern" refers to a technique for visually presenting the generated predicted verdict pattern to a user.

[0503] "Feedback" refers to the evaluation or additional information provided by the user in response to the predicted results.

[0504] "Retraining" refers to the process of retraining a machine learning model based on feedback from users.

[0505] MODE FOR CARRYING OUT THE INVENTION

[0506] The present invention is a system for providing predicted decisions on legal issues by utilizing a database of past legal cases. The system includes a means for a user to input legal issues, a means for managing the database of legal cases, a means for transmitting and receiving data, a means for analyzing data, a means for generating search queries, a means for searching legal case data, and a means for generating and presenting predicted decisions.

[0507] Hardware and Software Use

[0508] Terminal

[0509] The terminal provides an interface for users to enter legal information, for example, through a web browser or a dedicated application, and includes input fields where users can enter information using a keyboard or touchscreen.

[0510] server

[0511] The server receives the data entered by the user and parses it, using Python and NLP libraries (e.g., NLTK and spaCy) to parse the data and extract important keywords and phrases.

[0512] Based on the received data, a search query is generated for the case law database, and the server executes the generated query using a search engine such as Elasticsearch to retrieve relevant case law data.

[0513] The case data is preprocessed (using libraries such as Pandas and NumPy) and key features are extracted using natural language processing techniques (e.g., NLP models such as BERT).

[0514] Based on the extracted features, multiple predictive verdict patterns are generated using machine learning algorithms (using Scikit-learn and TensorFlow).

[0515] Prediction results

[0516] The server organizes the generated predicted verdict patterns and presents them in a user-friendly format. For example, the results are returned to the terminal in JSON format and rendered in HTML format on the terminal. The user can then view the results using a web browser or similar.

[0517] Feedback and model retraining

[0518] Users can provide feedback on the provided predictions, including a rating of the prediction's accuracy and additional information.

[0519] The server receives feedback and retrains the machine learning model as needed. The retraining process continuously improves the system's predictive accuracy.

[0520] Specific examples

[0521] For example, if a user writes "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," the server will analyze this information and search for similar past cases. The server extracts important features from the case data and applies a machine learning algorithm to present a predicted judgment, such as the following:

[0522] Order to pay all unpaid rent

[0523] "Order to pay repair costs and contract termination"

[0524] "Payment of 300,000 yen as settlement money and continuation of contract"

[0525] Example prompts for generative AI models

[0526] "Generate a predicted verdict based on past precedents for the following legal issue: a breach of lease agreement involving unpaid rent of 500,000 yen and unauthorized renovations during the lease term."

[0527] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0528] Step 1:

[0529] User Input Phase

[0530] The user uses the device interface to enter details of the legal issue, such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," into text input fields provided through a web browser or dedicated application.

[0531] Input: User-entered legal issue details text

[0532] Output: Structured data format (e.g. JSON) sent to the device

[0533] What happens: A web browser renders an HTML form, a user enters information using the keyboard, and clicks the submit button.

[0534] Step 2:

[0535] Data transmission

[0536] The device sends the legal details entered by the user to the server using the HTTPS protocol, encrypting the data.

[0537] Input: User-entered legal issue details

[0538] Output: Encrypted data packet sent to the server

[0539] Specific operation: The device generates an HTTP POST request and sends the data to the server, where the Web API receives the request.

[0540] Step 3:

[0541] Data Analysis and Query Generation

[0542] The server analyzes the received user-entered data, extracts key keywords and phrases, and generates a search query against the case law database.

[0543] Input: Legal issue details received from the device

[0544] Output: Search query against case law database

[0545] What it does: It uses Python and NLP libraries (e.g., NLTK or spaCy) on the server to extract important keywords and generate search queries based on those keywords.

[0546] Step 4:

[0547] Case Law Database Search

[0548] The server searches the case law database based on the generated search query to retrieve relevant cases, calculates a similarity score, and selects the most relevant case.

[0549] Input: search query

[0550] Output: Relevant case law data

[0551] What happens: The server uses a search engine such as Elasticsearch to issue a query and search the database.

[0552] Step 5:

[0553] Data Preprocessing

[0554] The server preprocesses the case data it receives, converting it into a format suitable for input to machine learning algorithms. Preprocessing includes text cleansing and normalization.

[0555] Input: Retrieved case data

[0556] Output: Preprocessed data format

[0557] Specific operation: Using libraries such as Pandas and NumPy on the server side, unnecessary information is removed and the data is normalized.

[0558] Step 6:

[0559] Feature extraction

[0560] The server uses natural language processing techniques to extract important features from the case data.

[0561] Input: Preprocessed case data

[0562] Output: Feature data

[0563] How it works: An NLP model (e.g., BERT) on the server analyzes the text data and extracts important features such as the issues at stake and the type of judgment.

[0564] Step 7:

[0565] Predictive Verdict Generation

[0566] The server generates multiple predicted verdict patterns using a machine learning algorithm based on the extracted features.

[0567] Input: Feature data

[0568] Output: Predicted verdict pattern

[0569] Specific operation: The server runs machine learning models using Scikit-learn and TensorFlow to generate predicted verdict patterns.

[0570] Step 8:

[0571] Prediction results

[0572] The server organizes the generated predicted verdict patterns and sends them to the terminal for presentation to the user. The results are displayed in a format that is easy for the user to understand.

[0573] Input: Predicted sentence pattern

[0574] Output: The resulting data that is presented to the user.

[0575] Specific operation: The result data is sent to the terminal in JSON format, converted to HTML format on the terminal, and displayed. The user can check the results using a web browser or similar.

[0576] Step 9:

[0577] Gathering feedback

[0578] The user can then provide feedback on the provided predictions, including a rating on the accuracy of the predictions and additional information.

[0579] Input: User feedback data

[0580] Output: Feedback data sent to the server

[0581] Specific behavior: The user enters feedback information using a web form on the device and clicks the submit button.

[0582] Step 10:

[0583] Retraining the Model

[0584] The server receives feedback from users and retrains the machine learning model as needed, improving prediction accuracy.

[0585] Input: Feedback data

[0586] Output: Retrained machine learning model

[0587] How it works: The server inputs the feedback data into the machine learning algorithm and retrains the model. The retraining process is performed periodically as a batch process.

[0588] (Application example 1)

[0589] 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."

[0590] Systems that provide predicted decisions for legal issues provide important information by utilizing databases of past legal precedents, but they have difficulty accurately assessing legal risks related to security risks. Another challenge is continually improving the accuracy of predicted decisions based on information entered by users. Furthermore, it is necessary to continuously improve accuracy by incorporating re-learning that reflects user feedback into the system. There is a need to solve these issues.

[0591] 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.

[0592] In this invention, the server includes means for a user to input a legal problem, means for managing a database of past legal cases, means for searching for similar legal cases based on the legal problem, means for extracting specific features from the search results, means for generating a plurality of predicted verdict patterns based on the extracted features, means for presenting the generated predicted verdict patterns to the user, means for inputting details of security risks, and means for performing legal risk assessment, thereby enabling users to predict verdicts for legal problems and security risks with high accuracy.

[0593] Definitions of important terms contained in the claims

[0594] "User" means any person or entity that utilizes the System to input legal issues and security risks and receive predicted adjudications and legal risk assessments.

[0595] A "legal issue" is a legal issue, dispute, or controversy that seeks a legal solution to a specific matter.

[0596] A "case law database" is a database that collects and organizes data on past trials, legal decisions, and verdicts, and is used to resolve similar legal issues.

[0597] "Similar cases" are past cases that have content very similar to the legal issue entered by the user.

[0598] "Features" refer to important elements and parameters extracted from the case law database, and serve as the basic data for predicting verdicts.

[0599] A "predicted judgment pattern" is a set of multiple judgment outcomes generated by a machine learning algorithm based on past case data for a given legal issue.

[0600] "Security risks" refer to threats and dangers to the security of companies and individuals, such as information leaks, system hacking, and internal crimes.

[0601] "Legal risk assessment" is the process of predicting and evaluating the legal consequences and impacts of security risks.

[0602] "Feedback" refers to opinions and suggestions for improvement regarding the use of the system provided by users, which are reflected in improving the accuracy of the system.

[0603] "Retraining" is the process of updating an existing machine learning model to improve its accuracy and performance based on user feedback and additional information.

[0604] MODE FOR CARRYING OUT THE INVENTION

[0605] In carrying out the present invention, a system for assessing legal issues and security risks is constructed, and a specific embodiment thereof is shown below.

[0606] First, the user uses a terminal to input details of legal issues and security risks. This input is done through a dedicated interface. For example, for legal issues, users can enter information such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovations during the contract period," while for security risks, users can enter details such as "Information leak," "System hacking," and "Employee internal crime."

[0607] The entered data is received and analyzed by the server. The server extracts important keywords and phrases from the text data and generates a search query based on these. The server manages a database of past legal precedents and uses the generated search query to quickly search for relevant legal precedents.

[0608] Next, the server extracts specific features from the search results. Natural language processing (NLP) technology is used to extract these features. Preprocessing is then performed, including text cleansing (removing unnecessary information and normalizing it) and conversion to structured data. Then, a machine learning algorithm is used to analyze the data based on the features, and multiple predictive verdict patterns are generated.

[0609] The generated predicted judgment patterns are presented to the user. Specific judgment outcomes, such as "order to pay all outstanding rent," "order to pay repair costs and terminate the contract," or "pay 300,000 yen as a settlement and continue the contract," are displayed as prediction results. Furthermore, a legal risk assessment of security risks is also presented, enabling companies and individuals to take appropriate countermeasures.

[0610] Users can input feedback on these prediction results on their devices, and the server retrains the machine learning model based on the received feedback. This retraining process continuously improves the accuracy of the system.

[0611] The specific implementation of this system uses the following hardware and software: a cloud server (e.g., AWS EC2, Google Cloud), a web framework (Flask), a machine learning library (Sklearn), and a library for saving and loading models (joblib).

[0612] For example, if a user inputs "a past case of a breach of a rental contract resulting in unpaid rent of 500,000 yen," the server will search the database of past legal precedents to extract similar cases. It will then present predicted judgment patterns such as "an order to pay the full amount of unpaid rent" or "an order to pay repair costs and terminate the contract," allowing the user to quickly obtain a legal decision.

[0613] Example prompt sentence:

[0614] "Unpaid rent of 500,000 yen, unauthorized renovations during the rental contract"

[0615] In this manner, the system of the present invention provides users with predicted verdicts and assessments of legal issues and security risks, providing legal decision support.

[0616] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0617] Processing steps of the system that realizes the application example

[0618] Step 1:

[0619] The user enters details about the legal issue or security risk.

[0620] The user enters the specific type of case and details of the problem (e.g., "Violation of rental contract" or "Unpaid rent: 500,000 yen") into the input field on the terminal.

[0621] Step 2:

[0622] The terminal transmits the input data to the server.

[0623] The terminal captures data through the input interface and transmits the data to the server using a secure communication protocol (such as HTTPS).

[0624] Step 3:

[0625] The server analyzes the received data and generates a search query.

[0626] The server uses natural language processing (NLP) techniques to extract important keywords and phrases from the text data, which then generate the search query.

[0627] Step 4:

[0628] The server searches a database of past legal precedents using the generated search query.

[0629] The server uses a high-speed search engine to search the database to identify similar cases, and retrieves the top N relevant cases based on the search criteria.

[0630] Step 5:

[0631] The server extracts specific features from the acquired case data.

[0632] The server preprocesses the data into a format suitable for machine learning algorithms, including text normalization and removal of unnecessary information, and extracts key features (e.g., issues at issue, type of decision, applicable law, sentencing, etc.).

[0633] Step 6:

[0634] The server generates a predicted verdict pattern based on the extracted features.

[0635] The server uses a machine learning model to generate multiple predictive judgment patterns using the extracted features as input. The model is a generative AI model that has learned from past case data.

[0636] Step 7:

[0637] The server presents the generated predicted sentence pattern to the user.

[0638] The server converts the prediction results into a user-friendly format and sends them to the terminal, where the user can check the predicted verdict results on the terminal interface.

[0639] Step 8:

[0640] The user inputs feedback on the provided prediction results.

[0641] The user inputs feedback about the accuracy of the prediction results and additional information into the terminal.

[0642] Step 9:

[0643] The server receives user feedback and retrains the system.

[0644] The server analyzes the feedback data and creates a retraining dataset, which improves the accuracy of the machine learning model.

[0645] In this way, the system provides users with predictive adjudications on legal issues and security risks, and continually improves its accuracy.

[0646] 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.

[0647] The present invention is a system that provides predicted decisions on legal issues by combining a database of past legal precedents with an emotion engine that identifies user emotions. The program processing of this system is explained below in natural language.

[0648] 1. User Input Phase

[0649] The device provides a user with an interface for inputting details of the legal issue and emotional state. The input fields include the type of case, specific details of the issue (e.g., amount of compensation, breach of contract), etc. It also includes sensors and question-type items for inputting or detecting the user's emotional state.

[0650] Through this interface, users can enter details of the legal issue they wish to discuss and, if necessary, their emotional state. For example, they can enter information such as "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," as well as their own emotional state (e.g., anger, anxiety).

[0651] The terminal transmits the data entered by the user and the detected emotional state to the server.

[0652] 2. Data reception and analysis by the server

[0653] The server receives the user's input data and emotion data transmitted from the terminal.

[0654] The server analyzes the received data and extracts the type of data (e.g., breach of contract, claim for compensation, etc.) and related parameters (e.g., outstanding amount, contract terms, etc.). It also analyzes the user's emotional state using an emotion engine.

[0655] 3. Database Search

[0656] The server generates a search query based on the analysis results, which is used to search for relevant past cases in a case law database.

[0657] The server uses the search query to rapidly search the case database to retrieve cases similar to the user's legal problem, resulting in the top N most relevant cases.

[0658] 4. Feature Extraction and Preprocessing

[0659] The server preprocesses the case data and converts it into a format suitable for machine learning algorithms. Preprocessing includes text cleansing (removing unnecessary information and normalizing) and feature extraction.

[0660] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, type of judgment, and sentence.

[0661] 5. Generating predicted verdict patterns

[0662] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the pre-processed and feature-extracted data and the user's emotional state. The machine learning model has already been trained based on past case data and can make highly accurate predictions for new data.

[0663] Predicted judgment patterns include, for example, "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and continuation of the contract."

[0664] 6. Presentation of results

[0665] The server selects the predicted judgment pattern that best suits the user's emotional state based on the emotion engine, thereby providing information in a form that is most acceptable to the user.

[0666] The server transmits the optimal predicted judgment pattern to the terminal.

[0667] The terminal displays the predicted verdict results sent from the server to the user, who can use this information to make legal decisions.

[0668] 7. Feedback and Retraining

[0669] The user can then provide feedback on the provided predictions, which may include a rating on the accuracy of the predictions or additional details.

[0670] The server receives user feedback and retrains the machine learning model as needed, continuously improving the system's prediction accuracy. This retraining also incorporates feedback based on the user's emotional state.

[0671] As a result, the present invention not only provides a predictive verdict for a user's legal problem, but also takes into account the user's emotional state to provide more appropriate advice in a fast and reliable manner. Specific implementation methods and processes are provided to help individuals and businesses effectively resolve legal issues.

[0672] The processing flow will be explained below.

[0673] Step 1: User Input

[0674] The terminal provides a user with an interface for entering details of the legal issue and emotional state, including fields for the type of case, details of the issue (e.g., amount of compensation, breach of contract).

[0675] Through the interface, users enter details of the legal issue they wish to discuss, and also enter their emotional state (e.g., anger, anxiety, etc.) in a field reflecting the emotional state.

[0676] The device sends the details of the legal issue and the emotional state entered by the user to the server in one batch.

[0677] Step 2: Receiving and analyzing data

[0678] The server receives the user's input data and emotion data transmitted from the terminal.

[0679] The server analyzes the received data and extracts the type of input data (e.g., breach of contract, claim for compensation, etc.) and related parameters (e.g., unpaid amount, contract terms, etc.). It also uses an emotion engine to analyze the user's emotional state.

[0680] Step 3: Search the database

[0681] The server generates a search query based on the analysis results, which includes specific keywords related to the law and emotional state.

[0682] The server uses the search query to quickly search the case law database, extracting cases similar to the user's legal problem, and retrieves the top N most relevant cases as search results.

[0683] Step 4: Feature extraction and preprocessing

[0684] The server analyzes the acquired case data and performs preprocessing, which includes text cleansing (removing unnecessary information and normalizing) and feature extraction.

[0685] The server uses natural language processing (NLP) technology to extract important features from case data, including the issues at issue, the type of judgment, and the sentence.

[0686] Step 5: Generate predicted verdict patterns

[0687] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the results of preprocessing and feature extraction and the user's emotional state. The algorithm is trained based on past case data and can make highly accurate predictions for new data.

[0688] Predicted judgment patterns are generated in specific forms, such as "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and the contract continue."

[0689] Step 6: Presenting the results

[0690] The server selects a predictive judgment pattern that is appropriate for the user's emotional state based on the emotion engine, and this selection provides information in a form that is most acceptable to the user.

[0691] The server transmits the optimal predicted judgment pattern to the terminal.

[0692] The terminal displays the predicted judgment results sent from the server to the user, who then makes legal decisions based on the displayed information.

[0693] Step 7: Processing feedback

[0694] The user can then input feedback on the presented prediction results, which may include an evaluation of the prediction accuracy or additional detailed information.

[0695] The server receives user feedback and retrains the machine learning model as needed, including feedback based on the user's emotional state, to continuously improve the system's prediction accuracy.

[0696] Through these specific processing steps, the present invention can provide a predicted judgment for the user's legal problem and further provide an optimal solution by taking the user's feelings into consideration.

[0697] Example 2

[0698] 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."

[0699] Conventional systems for legal issues are unable to provide predicted decisions that take into account the user's emotional state, making it difficult to provide optimal legal advice quickly. Furthermore, if the accuracy of the predicted decisions provided is low, there is no mechanism to improve the system's performance by incorporating user feedback, which limits the system's usefulness.

[0700] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for a user to input details of a legal issue and their emotional state, means for analyzing received data based on the legal issue and their emotional state, means for managing a database of past legal cases and searching for similar legal cases, means for extracting specific features from the search results and performing preprocessing, means for using a machine learning algorithm to generate multiple predicted verdict patterns based on the extracted features and the user's emotional state, means for selecting and presenting the generated predicted verdict patterns in an optimal form based on the user's emotional state, and means for collecting user feedback and performing re-learning to continuously improve the accuracy of the system. This makes it possible to quickly provide highly accurate predicted verdicts while taking the user's emotional state into consideration.

[0701] "Legal issue details" refers to the specific legal matter the user is attempting to address, including the type of matter and specific details of the issue (e.g., amount of compensation, details of contract breach).

[0702] "Emotional state" refers to the user's current mental or emotional state, and is information that expresses emotions such as "anger," "anxiety," and "sadness."

[0703] "Means for analysis" refers to the technical means for carrying out the process of analyzing input data and extracting data types and important parameters.

[0704] A "case law database" is a database system that systematically accumulates past legal decisions and court cases and manages them in a searchable format.

[0705] A "searching means" is a technical means for efficiently locating information in a database based on specific criteria.

[0706] "Features" are important attributes or metrics extracted from a dataset that machine learning models use to make predictions or classifications.

[0707] "Preprocessing" refers to processes such as data cleansing and normalization that transform data into a form suitable for analytics and machine learning algorithms.

[0708] A "machine learning algorithm" is a mathematical and statistical method for learning from data and making predictions or classifications.

[0709] A "predicted judgment pattern" is a pattern that indicates possible future court outcomes, generated based on analyzed data.

[0710] "Feedback" refers to user-provided evaluations and additional information about the system's predictive accuracy and usability.

[0711] "Retraining" is the process of updating a machine learning model and improving its performance based on new data and feedback collected.

[0712] The present invention is a system that provides predicted decisions for users' legal issues. The system combines a database of past legal cases with an emotion engine that identifies user emotions to provide more reliable legal advice.

[0713] The main components of the system are:

[0714] 1. User Interface - The terminal provides the user with an interface for entering details of the legal issue and emotional state. This interface includes fields for entering the case type, specific details of the issue, and emotional state.

[0715] 2. Data Reception and Analysis - The server receives the user's input data and emotion data sent from the device. The received data is analyzed using analysis software to extract the data type and important parameters. The emotion engine is also used to analyze the user's emotional state.

[0716] 3. Case database search - The server generates a search query based on the received data and analysis results, and searches the case database for the most relevant top N cases.

[0717] 4. Preprocessing and Feature Extraction - The server preprocesses the extracted case data, performing text cleansing and normalization, and then uses natural language processing (NLP) techniques to extract features such as the issues at issue, type of verdict, and sentence.

[0718] 5. Generation of predicted verdict patterns - The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the pre-processed data and the user's emotional state. This machine learning algorithm is trained based on past case data, allowing for highly accurate predictions.

[0719] 6. Presentation of results - The server selects the predicted verdict pattern that best suits the user's emotional state and sends it to the terminal. The terminal displays the received predicted verdict results to the user, allowing the user to make legal decisions based on this information.

[0720] 7. Feedback collection and retraining - The user inputs feedback on the provided predicted verdict results into the terminal. The server receives this feedback and retrains the machine learning model to improve the system's prediction accuracy.

[0721] Specific examples

[0722] For example, suppose a user wants to review a "violation of a rental contract." The user enters the following information into their terminal: "Unpaid rent: 500,000 yen," "Unauthorized renovation during the contract period," and selects "anger" as their emotional state. This information is sent to the server, where it is analyzed and searched through a precedent database. A predicted judgment pattern is generated based on the user's input and emotional state and presented to the user in the form of, for example, "Order to pay the full amount of unpaid rent," "Order to pay repair costs and terminate the contract," or "Pay 300,000 yen as a settlement and continue the contract."

[0723] Prompt Sentence Examples

[0724] An example of a prompt to input to a generative AI model is as follows:

[0725] text

[0726] User Input: Rental Agreement Violation

[0727] Problem details: Unpaid rent: 500,000 yen, unauthorized renovations during the contract period

[0728] Emotional state: Anger

[0729] System output: Judgment pattern 1: Order to pay the full amount of unpaid rent. Judgment pattern 2: Order to pay repair costs and terminate the contract. Judgment pattern 3: Payment of 300,000 yen as a settlement and continuation of the contract.

[0730] This prompt shows how the system will predict the outcome based on the information entered by the user. As can be seen from this example, the system takes into account the user's emotional state to provide the optimal predicted verdict, enabling it to provide fast and reliable legal advice.

[0731] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0732] Step 1: User Input Phase

[0733] The user inputs details of the legal issue and their emotional state into the terminal. Specifically, they input information such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period" through the user interface. They also select "Anger" as their emotional state. The input fields are equipped with text boxes and drop-down menus. This input data is temporarily stored in the terminal's memory and prepared for the next processing step.

[0734] Input: Details of the legal issue and emotional state entered by the user into the device

[0735] Output: User input data sent from the device to the server

[0736] Step 2: Data reception and analysis

[0737] The server receives user data sent from the device. This data is sent to the server in a structured data format such as JSON. The server uses data analysis software to analyze the input data and extract the type of legal issue and important parameters. It also uses an emotion engine to analyze the user's emotional state.

[0738] Input: User input data sent from the device to the server

[0739] Output: Analyzed legal issue type, parameters, and emotional state

[0740] Step 3: Search the database

[0741] The server generates a search query based on the analysis results. The generated query is used to search the case law database. For example, conditions such as "Violation of rental contract" and "Unpaid rent: 500,000 yen" are used as a query. The server sends this query to a database management system (DBMS) and quickly searches for related past cases.

[0742] Input: A search query generated based on the analysis results

[0743] Output: A set of relevant case law data

[0744] Step 4: Feature extraction and preprocessing

[0745] The server preprocesses the case data it receives. This preprocessing includes text cleansing and normalization. It uses natural language processing (NLP) techniques to extract features from the case data, such as the issues at issue, type of judgment, and sentence. This converts the data into a format suitable for machine learning algorithms.

[0746] Input: Searched case data

[0747] Output: Preprocessed and feature extracted data

[0748] Step 5: Generate predicted sentence patterns

[0749] The server uses a machine learning algorithm to generate multiple predictive verdict patterns based on the preprocessed and feature-extracted data and the user's emotional state. Because the machine learning model has already been trained on past case data, it can make highly accurate predictions for new data.

[0750] Input: Preprocessed and feature-extracted data, user emotional state

[0751] Output: A list of predicted verdict patterns

[0752] Step 6: Presenting the results

[0753] The server selects the predicted verdict pattern that best suits the user's emotional state from the generated patterns and sends it to the terminal. The terminal then displays the received predicted verdict results to the user, who can then make legal decisions based on this information.

[0754] Input: List of predicted sentence patterns

[0755] Output: The optimal predicted verdict pattern presented to the user.

[0756] Step 7: Feedback and Retraining

[0757] The user inputs feedback on the provided predicted verdict results into the terminal. The feedback includes an evaluation of the prediction accuracy and additional detailed information. The server receives this feedback and retrains the machine learning model. This retraining process continuously improves the system's prediction accuracy.

[0758] Input: User feedback

[0759] Output: The machine learning model after retraining

[0760] Through the above processing steps, the system can quickly provide highly accurate predictive verdicts that take into account the user's emotional state.

[0761] (Application example 2)

[0762] 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."

[0763] Conventional legal problem-solving systems only refer to past legal precedents for legal problems entered by users, and do not provide predicted verdicts that take the user's emotional state into account. This makes it difficult to provide predicted verdicts that are intuitive and easy for users to accept. Furthermore, while security operations require rapid and appropriate responses, conventional systems lack the functionality to support this.

[0764] 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.

[0765] In this invention, the server includes means for a user to input a legal issue and emotional state, means for managing a database of past legal cases, means for searching for similar legal cases based on the legal issue and emotional state, means for extracting specific features from the search results, means for generating a plurality of predicted verdict patterns based on the extracted features and the emotional state, means for presenting the generated predicted verdict patterns to the user, and means for selecting a predicted verdict pattern that best suits the user's emotional state, thereby making it possible to provide an effective and appropriate predicted verdict that takes into account the user's emotional state.

[0766] A "means for user input of legal issues and emotional state" is a device or software that provides an interface through which a user can input or detect details of a specific legal issue and their emotional state.

[0767] The "means for managing a database of past court cases" is a database system for storing past court case data and for efficiently searching and managing it.

[0768] The "means for searching for similar precedents based on the legal issue and emotional state" refers to an algorithm or technology that searches for relevant past precedents based on the legal issue and emotional state entered by the user.

[0769] "Means for extracting specific features from search results" refers to natural language processing and data analysis techniques for extracting important parameters and characteristics from the searched case law data.

[0770] The "means for generating a plurality of predicted verdict patterns based on the extracted features and the emotional state" refers to a machine learning model or algorithm that generates a plurality of possible predicted verdicts using the extracted data and the user's emotional state.

[0771] The "means for presenting the generated predicted judgment pattern to the user" refers to a screen or notification system that displays the predicted judgment result in an easy-to-understand manner for the user.

[0772] The "means for selecting a predicted verdict pattern that is most suitable for the user's emotional state" refers to an algorithm or system that takes into account the user's current emotional state and selects it to present the most appropriate predicted verdict.

[0773] "Means for improving the accuracy of generated predicted verdict patterns using a generative AI model" refers to technologies and methods that use a pre-trained generative AI model to improve the accuracy of the predicted verdicts provided.

[0774] The "means for receiving the user's feedback and re-training the generative AI model" refers to a method for improving prediction accuracy by collecting feedback provided by users and re-training the generative AI model based on that feedback.

[0775] This invention is a security operations support system that combines an interface for inputting users' legal issues and emotional state with a system for managing, searching, and analyzing past case law data. This system can be installed on smartphones, smart glasses, head-mounted displays, and robots. A specific embodiment of this system is described below.

[0776] Hardware and Software

[0777] Smartphone: Provides an interface for users to enter details of their legal issue and their emotional state.

[0778] Server: Receives and analyzes data, searches the case law database, extracts features, generates predicted judgment patterns, and presents the results to the user.

[0779] Natural Language Processing (NLP) technology: Used to extract specific features from retrieved cases.

[0780] Machine learning models: Used to generate multiple predictive verdict patterns based on extracted features and emotional states, including logistic regression and generative AI models.

[0781] TextBlob: A library used to parse emotional states from user input.

[0782] Database management system: Used to efficiently manage past case data.

[0783] Processing Step Description

[0784] 1. User Input Phase: The user inputs details of the specific legal problem and their emotional state through a smartphone interface. For example, they input the text, "I spotted a suspicious person while on night patrol. I was attacked by the person and temporarily hid. The situation was very dangerous, so I called the police." and their emotional state, "I was very anxious and confused."

[0785] 2. Data reception and analysis by the server: The server receives the data entered by the user and extracts features using natural language processing techniques (such as TfidfVectorizer). At the same time, it analyzes the emotional state using TextBlob.

[0786] 3. Database Search: The server searches similar past cases based on the legal issue and emotional state, thereby obtaining relevant cases that match the issue the user is facing.

[0787] 4. Feature extraction and preprocessing: Extract important features (e.g., issues at issue, type of judgment, sentence, etc.) from the retrieved cases and convert them into a format suitable for machine learning models (logistic regression, generative AI models, etc.).

[0788] 5. Generation of predicted judgment patterns: Based on the extracted features and emotional state, a machine learning model is used to generate multiple predicted judgment patterns, such as "Order to pay repair costs and terminate the contract," "Order to pay the full amount of unpaid rent," and "Pay 300,000 yen as a settlement and continue the contract."

[0789] 6. Presentation of results: The server selects the predicted judgment pattern that best suits the user's emotional state from the generated patterns and presents it to the user through the smartphone interface.

[0790] Specific prompt examples

[0791] As a concrete example, the system operates based on the following prompt sentence:

[0792] Case details: "While on night patrol, I spotted a suspicious individual. I was attacked by the individual and temporarily hid. The situation was extremely dangerous, so I called the police."

[0793] Emotional state: "Very anxious and confused."

[0794] Based on this prompt, the system generates an appropriate predictive verdict and provides users with appropriate and specific countermeasures, enabling quick and effective responses in the field of security operations.

[0795] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0796] Step 1: User Input Phase

[0797] Subject: User

[0798] Specific Actions: Users input details of specific legal issues and their emotional state through a smartphone interface.

[0799] Input: For example, the text "I spotted a suspicious person while on night patrol. I was attacked by the person and temporarily hid. The situation was very dangerous, so I called the police." and the emotional state "I was very anxious and confused."

[0800] Output: User input data (legal issue details and emotional state).

[0801] Step 2: Send data

[0802] Subject: Device

[0803] Specific operation: The terminal sends the data entered by the user to the server.

[0804] Input: Details of the legal issue and emotional state entered by the user.

[0805] Output: The user's input data received by the server.

[0806] Step 3: Data reception and analysis

[0807] Subject: Server

[0808] Specific operation: The server receives the user's input data sent from the device, analyzes the received data, extracts features using natural language processing techniques (such as TfidfVectorizer), and analyzes the emotional state using TextBlob.

[0809] Input: User input data sent by the device.

[0810] Output: Extracted features and parsed emotional states.

[0811] Step 4: Search the case law database

[0812] Subject: Server

[0813] Specific operation: The server searches for similar past cases based on the legal issue and emotional state, thereby obtaining relevant cases that match the issue the user is facing.

[0814] Input: Extracted features and parsed emotional states.

[0815] Output: Retrieved relevant case law data.

[0816] Step 5: Feature extraction and preprocessing

[0817] Subject: Server

[0818] Specific operation: The server extracts important features (e.g., issues at issue, type of judgment, sentence, etc.) from the retrieved case data and converts them into a format suitable for machine learning models (logistic regression, generative AI models, etc.).

[0819] Input: The relevant case law data retrieved.

[0820] Output: Preprocessed feature data.

[0821] Step 6: Generate predicted sentence patterns

[0822] Subject: Server

[0823] Specific operation: Based on the extracted features and emotional state, the server uses a machine learning model to generate multiple predicted judgment patterns, such as "order to pay repair costs and terminate the contract," "order to pay the full amount of unpaid rent," and "pay 300,000 yen as a settlement and continue the contract."

[0824] Input: Preprocessed feature data and parsed emotional states.

[0825] Output: Multiple predicted verdict patterns.

[0826] Step 7: Select and present results

[0827] Subject: Server

[0828] Specific operation: The server selects the predicted judgment pattern that best suits the user's emotional state from the generated predicted judgment patterns and presents it to the user through the smartphone interface.

[0829] Input: Multiple predicted judgment patterns and analyzed emotional states.

[0830] Output: The predicted verdict pattern displayed to the user.

[0831] 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.

[0832] 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.

[0833] 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.

[0834] [Third embodiment]

[0835] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0836] 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.

[0837] 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).

[0838] 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.

[0839] 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.

[0840] 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).

[0841] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0842] 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.

[0843] 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.

[0844] 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.

[0845] 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.

[0846] 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."

[0847] The present invention is a system that utilizes a database of past legal precedents to provide predicted decisions on legal issues. The program processing of this system will be explained below in natural language.

[0848] 1. User Input Phase

[0849] The terminal provides the user with an interface for entering details of the legal issue, including the type of case, specific details of the issue (e.g., amount of compensation, breach of contract).

[0850] Through this interface, users enter details of the legal issue they wish to discuss, such as "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovations during the contract period."

[0851] 2. Data reception and retrieval by the server

[0852] The terminal transmits the data entered by the user to the server.

[0853] The server parses the received user-entered data and generates a search query. Parsing involves the process of extracting important keywords and phrases from the textual input data.

[0854] The server quickly searches the case law database based on the generated search query, and retrieves the top N related cases based on the similarity score and search criteria.

[0855] 3. Feature Extraction and Prediction

[0856] The server preprocesses the case data it receives, converting it into a format suitable for input to machine learning algorithms. This preprocessing includes text cleansing (removing unnecessary information and normalizing) and converting it into structured data.

[0857] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, the type of judgment, applicable laws, and sentencing.

[0858] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the feature values. This algorithm learns from past case data as training data and makes highly accurate predictions for input data.

[0859] 4. Presenting the prediction results

[0860] The server extracts and organizes the generated predicted judgment patterns and converts them into a format that is easy for the user to understand. Specifically, it presents specific judgment results such as "order to pay the full amount of unpaid rent," "order to pay repair costs and terminate the contract," and "payment of 300,000 yen as a settlement and continuation of the contract."

[0861] The terminal displays this predicted judgment result to the user, who can refer to it and use it as information for making legal decisions.

[0862] 5. Feedback and Retraining

[0863] The user can then provide feedback on the prediction results, which may include rating the accuracy of the prediction or providing additional information.

[0864] The server receives user feedback and retrains the machine learning model as needed, continuously improving the accuracy of the system.

[0865] As described above, the present invention provides users with an efficient and reliable tool for quickly assessing legal issues and obtaining a predicted verdict. Through specific implementation methods and processes, the present invention can provide assistance to individuals and businesses in effectively resolving legal issues.

[0866] The processing flow will be explained below.

[0867] Step 1: User Input

[0868] The terminal provides the user with an interface for entering details of the legal matter, including fields for the type of case, the issues at stake, and the specific circumstances.

[0869] Through the interface, the user inputs details of the legal issue they wish to discuss, such as breach of lease agreement, unpaid rent of 500,000 yen, and unauthorized renovations during the lease term.

[0870] The terminal transmits the data entered by the user to the server.

[0871] Step 2: Receiving and analyzing data

[0872] The server receives the user input data sent from the terminal.

[0873] The server analyzes the received data and extracts the type of data (breach of contract, claim for compensation, etc.) and related parameters (amount due, contract terms, etc.).

[0874] Step 3: Search the database

[0875] The server generates a search query based on the extracted data, which is used to search for relevant past cases in a case law database.

[0876] The server uses the search query to quickly search through the case database to retrieve cases similar to the user's legal problem, resulting in the top N most relevant cases.

[0877] Step 4: Feature extraction and preprocessing

[0878] The server analyzes the acquired case data and preprocesses it into a format suitable for machine learning algorithms, including text cleansing (removing unnecessary information and normalizing it) and feature extraction.

[0879] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, type of judgment, and sentence.

[0880] Step 5: Generate predicted verdict patterns

[0881] The server inputs the pre-processed and feature-extracted data into a machine learning algorithm. The machine learning model has already been trained based on past case data and generates multiple predictive verdict patterns for new data.

[0882] Predicted judgment patterns include, for example, "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and continuation of the contract."

[0883] Step 6: Presenting the results

[0884] The server organizes the generated predicted judgment patterns and converts them into a format that is easy for the user to understand.

[0885] The server transmits the predicted judgment result to the terminal.

[0886] The terminal displays the predicted verdict results sent from the server to the user, who can use this information to make legal decisions.

[0887] Step 7: Processing feedback

[0888] The user can then input feedback on the presented prediction results, which may include an evaluation of the prediction accuracy or additional detailed information.

[0889] The server receives user feedback and retrains the machine learning model as needed, continuously improving the system's prediction accuracy.

[0890] These are the specific processing steps of the present invention, which allow individuals and businesses to obtain predicted decisions on legal matters in an efficient and reliable manner.

[0891] Example 1

[0892] 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."

[0893] Currently, there is a lack of systems that provide fast and reliable predictive decisions on legal issues. Many legal issues are complex, and researching past cases and making appropriate legal decisions requires a lot of time and expertise. This makes it difficult for individuals and businesses to receive support to effectively resolve legal issues. In addition, existing systems often lack the accuracy of their machine learning algorithms or retraining capabilities, resulting in inaccurate predictions.

[0894] 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.

[0895] In this invention, the server includes: a means for a user to input legal issues; a means for managing a database of past legal precedents; a means for transmitting user-input data to the server; a means for analyzing the received user-input data and generating a search query; a means for searching the legal precedent database and retrieving relevant legal precedents based on the generated search query; a means for preprocessing the legal precedent data and extracting important features using natural language processing technology; a means for generating multiple predicted verdict patterns using a machine learning algorithm based on the extracted features; and a means for presenting the generated predicted verdict patterns to the user. This enables the server to provide quick and accurate predicted verdicts for legal issues input by the user. Furthermore, the accuracy of predictions can be continuously improved by retraining the machine learning model based on user feedback.

[0896] "User" means any person or entity that utilizes the System to enter legal matters and obtain predicted judgments.

[0897] "Legal Issue" means detailed information about the case or matter that a User seeks to resolve.

[0898] "Means for input" refers to an interface that allows a user to enter details of a legal issue in text form.

[0899] "Means for managing a case law database" refers to technologies and systems that store, search, and manage past case law information in a format that can be referenced.

[0900] "Means for transmitting data to a server" refers to the protocols and technologies used to transmit information entered by a user from a terminal to a server.

[0901] "Means for analyzing data" refers to technology used to process received user-entered data and extract required keywords and phrases.

[0902] "Means for generating a search query" refers to a technology for generating a query for searching a case law database based on the analyzed data.

[0903] The "means for searching a case law database" refers to a technology for quickly searching a case law database using a generated query to obtain relevant cases.

[0904] "Means for pre-processing data" refers to the process of preparing acquired case law data in a format suitable for input into machine learning algorithms.

[0905] "Means for extracting features" refers to technology that extracts important elements from case law data and organizes them into a format that can be used in machine learning models.

[0906] "Means for generating predicted verdict patterns" refers to the process of creating multiple predicted verdict patterns using a machine learning algorithm based on the extracted features.

[0907] The "means for presenting a predicted verdict pattern" refers to a technique for visually presenting the generated predicted verdict pattern to a user.

[0908] "Feedback" refers to the evaluation or additional information provided by the user in response to the predicted results.

[0909] "Retraining" refers to the process of retraining a machine learning model based on feedback from users.

[0910] MODE FOR CARRYING OUT THE INVENTION

[0911] The present invention is a system for providing predicted decisions on legal issues by utilizing a database of past legal cases. The system includes a means for a user to input legal issues, a means for managing the database of legal cases, a means for transmitting and receiving data, a means for analyzing data, a means for generating search queries, a means for searching legal case data, and a means for generating and presenting predicted decisions.

[0912] Hardware and Software Use

[0913] Terminal

[0914] The terminal provides an interface for users to enter legal information, for example, through a web browser or a dedicated application, and includes input fields where users can enter information using a keyboard or touchscreen.

[0915] server

[0916] The server receives the data entered by the user and parses it, using Python and NLP libraries (e.g., NLTK and spaCy) to parse the data and extract important keywords and phrases.

[0917] Based on the received data, a search query is generated for the case law database, and the server executes the generated query using a search engine such as Elasticsearch to retrieve relevant case law data.

[0918] The case data is preprocessed (using libraries such as Pandas and NumPy) and key features are extracted using natural language processing techniques (e.g., NLP models such as BERT).

[0919] Based on the extracted features, multiple predictive verdict patterns are generated using machine learning algorithms (using Scikit-learn and TensorFlow).

[0920] Prediction results

[0921] The server organizes the generated predicted verdict patterns and presents them in a user-friendly format. For example, the results are returned to the terminal in JSON format and rendered in HTML format on the terminal. The user can then view the results using a web browser or similar.

[0922] Feedback and model retraining

[0923] Users can provide feedback on the provided predictions, including a rating of the prediction's accuracy and additional information.

[0924] The server receives feedback and retrains the machine learning model as needed. The retraining process continuously improves the system's predictive accuracy.

[0925] Specific examples

[0926] For example, if a user writes "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," the server will analyze this information and search for similar past cases. The server extracts important features from the case data and applies a machine learning algorithm to present a predicted judgment, such as the following:

[0927] Order to pay all unpaid rent

[0928] "Order to pay repair costs and contract termination"

[0929] "Payment of 300,000 yen as settlement money and continuation of contract"

[0930] Example prompts for generative AI models

[0931] "Generate a predicted verdict based on past precedents for the following legal issue: a breach of lease agreement involving unpaid rent of 500,000 yen and unauthorized renovations during the lease term."

[0932] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0933] Step 1:

[0934] User Input Phase

[0935] The user uses the device interface to enter details of the legal issue, such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," into text input fields provided through a web browser or dedicated application.

[0936] Input: User-entered legal issue details text

[0937] Output: Structured data format (e.g. JSON) sent to the device

[0938] What happens: A web browser renders an HTML form, a user enters information using the keyboard, and clicks the submit button.

[0939] Step 2:

[0940] Data transmission

[0941] The device sends the legal details entered by the user to the server using the HTTPS protocol, encrypting the data.

[0942] Input: User-entered legal issue details

[0943] Output: Encrypted data packet sent to the server

[0944] Specific operation: The device generates an HTTP POST request and sends the data to the server, where the Web API receives the request.

[0945] Step 3:

[0946] Data Analysis and Query Generation

[0947] The server analyzes the received user-entered data, extracts key keywords and phrases, and generates a search query against the case law database.

[0948] Input: Legal issue details received from the device

[0949] Output: Search query against case law database

[0950] What it does: It uses Python and NLP libraries (e.g., NLTK or spaCy) on the server to extract important keywords and generate search queries based on those keywords.

[0951] Step 4:

[0952] Case Law Database Search

[0953] The server searches the case law database based on the generated search query to retrieve relevant cases, calculates a similarity score, and selects the most relevant case.

[0954] Input: search query

[0955] Output: Relevant case law data

[0956] What happens: The server uses a search engine such as Elasticsearch to issue a query and search the database.

[0957] Step 5:

[0958] Data Preprocessing

[0959] The server preprocesses the case data it receives, converting it into a format suitable for input to machine learning algorithms. Preprocessing includes text cleansing and normalization.

[0960] Input: Retrieved case data

[0961] Output: Preprocessed data format

[0962] Specific operation: Using libraries such as Pandas and NumPy on the server side, unnecessary information is removed and the data is normalized.

[0963] Step 6:

[0964] Feature extraction

[0965] The server uses natural language processing techniques to extract important features from the case data.

[0966] Input: Preprocessed case data

[0967] Output: Feature data

[0968] How it works: An NLP model (e.g., BERT) on the server analyzes the text data and extracts important features such as the issues at stake and the type of judgment.

[0969] Step 7:

[0970] Predictive Verdict Generation

[0971] The server generates multiple predicted verdict patterns using a machine learning algorithm based on the extracted features.

[0972] Input: Feature data

[0973] Output: Predicted verdict pattern

[0974] Specific operation: The server runs machine learning models using Scikit-learn and TensorFlow to generate predicted verdict patterns.

[0975] Step 8:

[0976] Prediction results

[0977] The server organizes the generated predicted verdict patterns and sends them to the terminal for presentation to the user. The results are displayed in a format that is easy for the user to understand.

[0978] Input: Predicted sentence pattern

[0979] Output: The resulting data that is presented to the user.

[0980] Specific operation: The result data is sent to the terminal in JSON format, converted to HTML format on the terminal, and displayed. The user can check the results using a web browser or similar.

[0981] Step 9:

[0982] Gathering feedback

[0983] The user can then provide feedback on the provided predictions, including a rating on the accuracy of the predictions and additional information.

[0984] Input: User feedback data

[0985] Output: Feedback data sent to the server

[0986] Specific behavior: The user enters feedback information using a web form on the device and clicks the submit button.

[0987] Step 10:

[0988] Retraining the Model

[0989] The server receives feedback from users and retrains the machine learning model as needed, improving prediction accuracy.

[0990] Input: Feedback data

[0991] Output: Retrained machine learning model

[0992] How it works: The server inputs the feedback data into the machine learning algorithm and retrains the model. The retraining process is performed periodically as a batch process.

[0993] (Application example 1)

[0994] 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."

[0995] Systems that provide predicted decisions for legal issues provide important information by utilizing databases of past legal precedents, but they have difficulty accurately assessing legal risks related to security risks. Another challenge is continually improving the accuracy of predicted decisions based on information entered by users. Furthermore, it is necessary to continuously improve accuracy by incorporating re-learning that reflects user feedback into the system. There is a need to solve these issues.

[0996] 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.

[0997] In this invention, the server includes means for a user to input a legal problem, means for managing a database of past legal cases, means for searching for similar legal cases based on the legal problem, means for extracting specific features from the search results, means for generating a plurality of predicted verdict patterns based on the extracted features, means for presenting the generated predicted verdict patterns to the user, means for inputting details of security risks, and means for performing legal risk assessment, thereby enabling users to predict verdicts for legal problems and security risks with high accuracy.

[0998] Definitions of important terms contained in the claims

[0999] "User" means any person or entity that utilizes the System to input legal issues and security risks and receive predicted adjudications and legal risk assessments.

[1000] A "legal issue" is a legal issue, dispute, or controversy that seeks a legal solution to a specific matter.

[1001] A "case law database" is a database that collects and organizes data on past trials, legal decisions, and verdicts, and is used to resolve similar legal issues.

[1002] "Similar cases" are past cases that have content very similar to the legal issue entered by the user.

[1003] "Features" refer to important elements and parameters extracted from the case law database, and serve as the basic data for predicting verdicts.

[1004] A "predicted judgment pattern" is a set of multiple judgment outcomes generated by a machine learning algorithm based on past case data for a given legal issue.

[1005] "Security risks" refer to threats and dangers to the security of companies and individuals, such as information leaks, system hacking, and internal crimes.

[1006] "Legal risk assessment" is the process of predicting and evaluating the legal consequences and impacts of security risks.

[1007] "Feedback" refers to opinions and suggestions for improvement regarding the use of the system provided by users, which are reflected in improving the accuracy of the system.

[1008] "Retraining" is the process of updating an existing machine learning model to improve its accuracy and performance based on user feedback and additional information.

[1009] MODE FOR CARRYING OUT THE INVENTION

[1010] In carrying out the present invention, a system for assessing legal issues and security risks is constructed, and a specific embodiment thereof is shown below.

[1011] First, the user uses a terminal to input details of legal issues and security risks. This input is done through a dedicated interface. For example, for legal issues, users can enter information such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovations during the contract period," while for security risks, users can enter details such as "Information leak," "System hacking," and "Employee internal crime."

[1012] The entered data is received and analyzed by the server. The server extracts important keywords and phrases from the text data and generates a search query based on these. The server manages a database of past legal precedents and uses the generated search query to quickly search for relevant legal precedents.

[1013] Next, the server extracts specific features from the search results. Natural language processing (NLP) technology is used to extract these features. Preprocessing is then performed, including text cleansing (removing unnecessary information and normalizing it) and conversion to structured data. Then, a machine learning algorithm is used to analyze the data based on the features, and multiple predictive verdict patterns are generated.

[1014] The generated predicted judgment patterns are presented to the user. Specific judgment outcomes, such as "order to pay all outstanding rent," "order to pay repair costs and terminate the contract," or "pay 300,000 yen as a settlement and continue the contract," are displayed as prediction results. Furthermore, a legal risk assessment of security risks is also presented, enabling companies and individuals to take appropriate countermeasures.

[1015] Users can input feedback on these prediction results on their devices, and the server retrains the machine learning model based on the received feedback. This retraining process continuously improves the accuracy of the system.

[1016] The specific implementation of this system uses the following hardware and software: a cloud server (e.g., AWS EC2, Google Cloud), a web framework (Flask), a machine learning library (Sklearn), and a library for saving and loading models (joblib).

[1017] For example, if a user inputs "a past case of a breach of a rental contract resulting in unpaid rent of 500,000 yen," the server will search the database of past legal precedents to extract similar cases. It will then present predicted judgment patterns such as "an order to pay the full amount of unpaid rent" or "an order to pay repair costs and terminate the contract," allowing the user to quickly obtain a legal decision.

[1018] Example prompt sentence:

[1019] "Unpaid rent of 500,000 yen, unauthorized renovations during the rental contract"

[1020] In this manner, the system of the present invention provides users with predicted verdicts and assessments of legal issues and security risks, providing legal decision support.

[1021] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1022] Processing steps of the system that realizes the application example

[1023] Step 1:

[1024] The user enters details about the legal issue or security risk.

[1025] The user enters the specific type of case and details of the problem (e.g., "Violation of rental contract" or "Unpaid rent: 500,000 yen") into the input field on the terminal.

[1026] Step 2:

[1027] The terminal transmits the input data to the server.

[1028] The terminal captures data through the input interface and transmits the data to the server using a secure communication protocol (such as HTTPS).

[1029] Step 3:

[1030] The server analyzes the received data and generates a search query.

[1031] The server uses natural language processing (NLP) techniques to extract important keywords and phrases from the text data, which then generate the search query.

[1032] Step 4:

[1033] The server searches a database of past legal precedents using the generated search query.

[1034] The server uses a high-speed search engine to search the database to identify similar cases, and retrieves the top N relevant cases based on the search criteria.

[1035] Step 5:

[1036] The server extracts specific features from the acquired case data.

[1037] The server preprocesses the data into a format suitable for machine learning algorithms, including text normalization and removal of unnecessary information, and extracts key features (e.g., issues at issue, type of decision, applicable law, sentencing, etc.).

[1038] Step 6:

[1039] The server generates a predicted verdict pattern based on the extracted features.

[1040] The server uses a machine learning model to generate multiple predictive judgment patterns using the extracted features as input. The model is a generative AI model that has learned from past case data.

[1041] Step 7:

[1042] The server presents the generated predicted sentence pattern to the user.

[1043] The server converts the prediction results into a user-friendly format and sends them to the terminal, where the user can check the predicted verdict results on the terminal interface.

[1044] Step 8:

[1045] The user inputs feedback on the provided prediction results.

[1046] The user inputs feedback about the accuracy of the prediction results and additional information into the terminal.

[1047] Step 9:

[1048] The server receives user feedback and retrains the system.

[1049] The server analyzes the feedback data and creates a retraining dataset, which improves the accuracy of the machine learning model.

[1050] In this way, the system provides users with predictive adjudications on legal issues and security risks, and continually improves its accuracy.

[1051] 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.

[1052] The present invention is a system that provides predicted decisions on legal issues by combining a database of past legal precedents with an emotion engine that identifies user emotions. The program processing of this system is explained below in natural language.

[1053] 1. User Input Phase

[1054] The device provides a user with an interface for inputting details of the legal issue and emotional state. The input fields include the type of case, specific details of the issue (e.g., amount of compensation, breach of contract), etc. It also includes sensors and question-type items for inputting or detecting the user's emotional state.

[1055] Through this interface, users can enter details of the legal issue they wish to discuss and, if necessary, their emotional state. For example, they can enter information such as "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," as well as their own emotional state (e.g., anger, anxiety).

[1056] The terminal transmits the data entered by the user and the detected emotional state to the server.

[1057] 2. Data reception and analysis by the server

[1058] The server receives the user's input data and emotion data transmitted from the terminal.

[1059] The server analyzes the received data and extracts the type of data (e.g., breach of contract, claim for compensation, etc.) and related parameters (e.g., outstanding amount, contract terms, etc.). It also analyzes the user's emotional state using an emotion engine.

[1060] 3. Database Search

[1061] The server generates a search query based on the analysis results, which is used to search for relevant past cases in a case law database.

[1062] The server uses the search query to rapidly search the case database to retrieve cases similar to the user's legal problem, resulting in the top N most relevant cases.

[1063] 4. Feature Extraction and Preprocessing

[1064] The server preprocesses the case data and converts it into a format suitable for machine learning algorithms. Preprocessing includes text cleansing (removing unnecessary information and normalizing) and feature extraction.

[1065] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, type of judgment, and sentence.

[1066] 5. Generating predicted verdict patterns

[1067] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the pre-processed and feature-extracted data and the user's emotional state. The machine learning model has already been trained based on past case data and can make highly accurate predictions for new data.

[1068] Predicted judgment patterns include, for example, "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and continuation of the contract."

[1069] 6. Presentation of results

[1070] The server selects the predicted judgment pattern that best suits the user's emotional state based on the emotion engine, thereby providing information in a form that is most acceptable to the user.

[1071] The server transmits the optimal predicted judgment pattern to the terminal.

[1072] The terminal displays the predicted verdict results sent from the server to the user, who can use this information to make legal decisions.

[1073] 7. Feedback and Retraining

[1074] The user can then provide feedback on the provided predictions, which may include a rating on the accuracy of the predictions or additional details.

[1075] The server receives user feedback and retrains the machine learning model as needed, continuously improving the system's prediction accuracy. This retraining also incorporates feedback based on the user's emotional state.

[1076] As a result, the present invention not only provides a predictive verdict for a user's legal problem, but also takes into account the user's emotional state to provide more appropriate advice in a fast and reliable manner. Specific implementation methods and processes are provided to help individuals and businesses effectively resolve legal issues.

[1077] The processing flow will be explained below.

[1078] Step 1: User Input

[1079] The terminal provides a user with an interface for entering details of the legal issue and emotional state, including fields for the type of case, details of the issue (e.g., amount of compensation, breach of contract).

[1080] Through the interface, users enter details of the legal issue they wish to discuss, and also enter their emotional state (e.g., anger, anxiety, etc.) in a field reflecting the emotional state.

[1081] The device sends the details of the legal issue and the emotional state entered by the user to the server in one batch.

[1082] Step 2: Receiving and analyzing data

[1083] The server receives the user's input data and emotion data transmitted from the terminal.

[1084] The server analyzes the received data and extracts the type of input data (e.g., breach of contract, claim for compensation, etc.) and related parameters (e.g., unpaid amount, contract terms, etc.). It also uses an emotion engine to analyze the user's emotional state.

[1085] Step 3: Search the database

[1086] The server generates a search query based on the analysis results, which includes specific keywords related to the law and emotional state.

[1087] The server uses the search query to quickly search the case law database, extracting cases similar to the user's legal problem, and retrieves the top N most relevant cases as search results.

[1088] Step 4: Feature extraction and preprocessing

[1089] The server analyzes the acquired case data and performs preprocessing, which includes text cleansing (removing unnecessary information and normalizing) and feature extraction.

[1090] The server uses natural language processing (NLP) technology to extract important features from case data, including the issues at issue, the type of judgment, and the sentence.

[1091] Step 5: Generate predicted verdict patterns

[1092] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the results of preprocessing and feature extraction and the user's emotional state. The algorithm is trained based on past case data and can make highly accurate predictions for new data.

[1093] Predicted judgment patterns are generated in specific forms, such as "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and the contract continue."

[1094] Step 6: Presenting the results

[1095] The server selects a predictive judgment pattern that is appropriate for the user's emotional state based on the emotion engine, and this selection provides information in a form that is most acceptable to the user.

[1096] The server transmits the optimal predicted judgment pattern to the terminal.

[1097] The terminal displays the predicted judgment results sent from the server to the user, who then makes legal decisions based on the displayed information.

[1098] Step 7: Processing feedback

[1099] The user can then input feedback on the presented prediction results, which may include an evaluation of the prediction accuracy or additional detailed information.

[1100] The server receives user feedback and retrains the machine learning model as needed, including feedback based on the user's emotional state, to continuously improve the system's prediction accuracy.

[1101] Through these specific processing steps, the present invention can provide a predicted judgment for the user's legal problem and further provide an optimal solution by taking the user's feelings into consideration.

[1102] Example 2

[1103] 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."

[1104] Conventional systems for legal issues are unable to provide predicted decisions that take into account the user's emotional state, making it difficult to provide optimal legal advice quickly. Furthermore, if the accuracy of the predicted decisions provided is low, there is no mechanism to improve the system's performance by incorporating user feedback, which limits the system's usefulness.

[1105] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for a user to input details of a legal issue and their emotional state, means for analyzing received data based on the legal issue and their emotional state, means for managing a database of past legal cases and searching for similar legal cases, means for extracting specific features from the search results and performing preprocessing, means for using a machine learning algorithm to generate multiple predicted verdict patterns based on the extracted features and the user's emotional state, means for selecting and presenting the generated predicted verdict patterns in an optimal form based on the user's emotional state, and means for collecting user feedback and performing re-learning to continuously improve the accuracy of the system. This makes it possible to quickly provide highly accurate predicted verdicts while taking the user's emotional state into consideration.

[1106] "Legal issue details" refers to the specific legal matter the user is attempting to address, including the type of matter and specific details of the issue (e.g., amount of compensation, details of contract breach).

[1107] "Emotional state" refers to the user's current mental or emotional state, and is information that expresses emotions such as "anger," "anxiety," and "sadness."

[1108] "Means for analysis" refers to the technical means for carrying out the process of analyzing input data and extracting data types and important parameters.

[1109] A "case law database" is a database system that systematically accumulates past legal decisions and court cases and manages them in a searchable format.

[1110] A "searching means" is a technical means for efficiently locating information in a database based on specific criteria.

[1111] "Features" are important attributes or metrics extracted from a dataset that machine learning models use to make predictions or classifications.

[1112] "Preprocessing" refers to processes such as data cleansing and normalization that transform data into a form suitable for analytics and machine learning algorithms.

[1113] A "machine learning algorithm" is a mathematical and statistical method for learning from data and making predictions or classifications.

[1114] A "predicted judgment pattern" is a pattern that indicates possible future court outcomes, generated based on analyzed data.

[1115] "Feedback" refers to user-provided evaluations and additional information about the system's predictive accuracy and usability.

[1116] "Retraining" is the process of updating a machine learning model and improving its performance based on new data and feedback collected.

[1117] The present invention is a system that provides predicted decisions for users' legal issues. The system combines a database of past legal cases with an emotion engine that identifies user emotions to provide more reliable legal advice.

[1118] The main components of the system are:

[1119] 1. User Interface - The terminal provides the user with an interface for entering details of the legal issue and emotional state. This interface includes fields for entering the case type, specific details of the issue, and emotional state.

[1120] 2. Data Reception and Analysis - The server receives the user's input data and emotion data sent from the device. The received data is analyzed using analysis software to extract the data type and important parameters. The emotion engine is also used to analyze the user's emotional state.

[1121] 3. Case database search - The server generates a search query based on the received data and analysis results, and searches the case database for the most relevant top N cases.

[1122] 4. Preprocessing and Feature Extraction - The server preprocesses the extracted case data, performing text cleansing and normalization, and then uses natural language processing (NLP) techniques to extract features such as the issues at issue, type of verdict, and sentence.

[1123] 5. Generation of predicted verdict patterns - The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the pre-processed data and the user's emotional state. This machine learning algorithm is trained based on past case data, allowing for highly accurate predictions.

[1124] 6. Presentation of results - The server selects the predicted verdict pattern that best suits the user's emotional state and sends it to the terminal. The terminal displays the received predicted verdict results to the user, allowing the user to make legal decisions based on this information.

[1125] 7. Feedback collection and retraining - The user inputs feedback on the provided predicted verdict results into the terminal. The server receives this feedback and retrains the machine learning model to improve the system's prediction accuracy.

[1126] Specific examples

[1127] For example, suppose a user wants to review a "violation of a rental contract." The user enters the following information into their terminal: "Unpaid rent: 500,000 yen," "Unauthorized renovation during the contract period," and selects "anger" as their emotional state. This information is sent to the server, where it is analyzed and searched through a precedent database. A predicted judgment pattern is generated based on the user's input and emotional state and presented to the user in the form of, for example, "Order to pay the full amount of unpaid rent," "Order to pay repair costs and terminate the contract," or "Pay 300,000 yen as a settlement and continue the contract."

[1128] Prompt Sentence Examples

[1129] An example of a prompt to input to a generative AI model is as follows:

[1130] text

[1131] User Input: Rental Agreement Violation

[1132] Problem details: Unpaid rent: 500,000 yen, unauthorized renovations during the contract period

[1133] Emotional state: Anger

[1134] System output: Judgment pattern 1: Order to pay the full amount of unpaid rent. Judgment pattern 2: Order to pay repair costs and terminate the contract. Judgment pattern 3: Payment of 300,000 yen as a settlement and continuation of the contract.

[1135] This prompt shows how the system will predict the outcome based on the information entered by the user. As can be seen from this example, the system takes into account the user's emotional state to provide the optimal predicted verdict, enabling it to provide fast and reliable legal advice.

[1136] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1137] Step 1: User Input Phase

[1138] The user inputs details of the legal issue and their emotional state into the terminal. Specifically, they input information such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period" through the user interface. They also select "Anger" as their emotional state. The input fields are equipped with text boxes and drop-down menus. This input data is temporarily stored in the terminal's memory and prepared for the next processing step.

[1139] Input: Details of the legal issue and emotional state entered by the user into the device

[1140] Output: User input data sent from the device to the server

[1141] Step 2: Data reception and analysis

[1142] The server receives user data sent from the device. This data is sent to the server in a structured data format such as JSON. The server uses data analysis software to analyze the input data and extract the type of legal issue and important parameters. It also uses an emotion engine to analyze the user's emotional state.

[1143] Input: User input data sent from the device to the server

[1144] Output: Analyzed legal issue type, parameters, and emotional state

[1145] Step 3: Search the database

[1146] The server generates a search query based on the analysis results. The generated query is used to search the case law database. For example, conditions such as "Violation of rental contract" and "Unpaid rent: 500,000 yen" are used as a query. The server sends this query to a database management system (DBMS) and quickly searches for related past cases.

[1147] Input: A search query generated based on the analysis results

[1148] Output: A set of relevant case law data

[1149] Step 4: Feature extraction and preprocessing

[1150] The server preprocesses the case data it receives. This preprocessing includes text cleansing and normalization. It uses natural language processing (NLP) techniques to extract features from the case data, such as the issues at issue, type of judgment, and sentence. This converts the data into a format suitable for machine learning algorithms.

[1151] Input: Searched case data

[1152] Output: Preprocessed and feature extracted data

[1153] Step 5: Generate predicted sentence patterns

[1154] The server uses a machine learning algorithm to generate multiple predictive verdict patterns based on the preprocessed and feature-extracted data and the user's emotional state. Because the machine learning model has already been trained on past case data, it can make highly accurate predictions for new data.

[1155] Input: Preprocessed and feature-extracted data, user emotional state

[1156] Output: A list of predicted verdict patterns

[1157] Step 6: Presenting the results

[1158] The server selects the predicted verdict pattern that best suits the user's emotional state from the generated patterns and sends it to the terminal. The terminal then displays the received predicted verdict results to the user, who can then make legal decisions based on this information.

[1159] Input: List of predicted sentence patterns

[1160] Output: The optimal predicted verdict pattern presented to the user.

[1161] Step 7: Feedback and Retraining

[1162] The user inputs feedback on the provided predicted verdict results into the terminal. The feedback includes an evaluation of the prediction accuracy and additional detailed information. The server receives this feedback and retrains the machine learning model. This retraining process continuously improves the system's prediction accuracy.

[1163] Input: User feedback

[1164] Output: The machine learning model after retraining

[1165] Through the above processing steps, the system can quickly provide highly accurate predictive verdicts that take into account the user's emotional state.

[1166] (Application example 2)

[1167] 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."

[1168] Conventional legal problem-solving systems only refer to past legal precedents for legal problems entered by users, and do not provide predicted verdicts that take the user's emotional state into account. This makes it difficult to provide predicted verdicts that are intuitive and easy for users to accept. Furthermore, while security operations require rapid and appropriate responses, conventional systems lack the functionality to support this.

[1169] 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.

[1170] In this invention, the server includes means for a user to input a legal issue and emotional state, means for managing a database of past legal cases, means for searching for similar legal cases based on the legal issue and emotional state, means for extracting specific features from the search results, means for generating a plurality of predicted verdict patterns based on the extracted features and the emotional state, means for presenting the generated predicted verdict patterns to the user, and means for selecting a predicted verdict pattern that best suits the user's emotional state, thereby making it possible to provide an effective and appropriate predicted verdict that takes into account the user's emotional state.

[1171] A "means for user input of legal issues and emotional state" is a device or software that provides an interface through which a user can input or detect details of a specific legal issue and their emotional state.

[1172] The "means for managing a database of past court cases" is a database system for storing past court case data and for efficiently searching and managing it.

[1173] The "means for searching for similar precedents based on the legal issue and emotional state" refers to an algorithm or technology that searches for relevant past precedents based on the legal issue and emotional state entered by the user.

[1174] "Means for extracting specific features from search results" refers to natural language processing and data analysis techniques for extracting important parameters and characteristics from the searched case law data.

[1175] The "means for generating a plurality of predicted verdict patterns based on the extracted features and the emotional state" refers to a machine learning model or algorithm that generates a plurality of possible predicted verdicts using the extracted data and the user's emotional state.

[1176] The "means for presenting the generated predicted judgment pattern to the user" refers to a screen or notification system that displays the predicted judgment result in an easy-to-understand manner for the user.

[1177] The "means for selecting a predicted verdict pattern that is most suitable for the user's emotional state" refers to an algorithm or system that takes into account the user's current emotional state and selects it to present the most appropriate predicted verdict.

[1178] "Means for improving the accuracy of generated predicted verdict patterns using a generative AI model" refers to technologies and methods that use a pre-trained generative AI model to improve the accuracy of the predicted verdicts provided.

[1179] The "means for receiving the user's feedback and re-training the generative AI model" refers to a method for improving prediction accuracy by collecting feedback provided by users and re-training the generative AI model based on that feedback.

[1180] This invention is a security operations support system that combines an interface for inputting users' legal issues and emotional state with a system for managing, searching, and analyzing past case law data. This system can be installed on smartphones, smart glasses, head-mounted displays, and robots. A specific embodiment of this system is described below.

[1181] Hardware and Software

[1182] Smartphone: Provides an interface for users to enter details of their legal issue and their emotional state.

[1183] Server: Receives and analyzes data, searches the case law database, extracts features, generates predicted judgment patterns, and presents the results to the user.

[1184] Natural Language Processing (NLP) technology: Used to extract specific features from retrieved cases.

[1185] Machine learning models: Used to generate multiple predictive verdict patterns based on extracted features and emotional states, including logistic regression and generative AI models.

[1186] TextBlob: A library used to parse emotional states from user input.

[1187] Database management system: Used to efficiently manage past case data.

[1188] Processing Step Description

[1189] 1. User Input Phase: The user inputs details of the specific legal problem and their emotional state through a smartphone interface. For example, they input the text, "I spotted a suspicious person while on night patrol. I was attacked by the person and temporarily hid. The situation was very dangerous, so I called the police." and their emotional state, "I was very anxious and confused."

[1190] 2. Data reception and analysis by the server: The server receives the data entered by the user and extracts features using natural language processing techniques (such as TfidfVectorizer). At the same time, it analyzes the emotional state using TextBlob.

[1191] 3. Database Search: The server searches similar past cases based on the legal issue and emotional state, thereby obtaining relevant cases that match the issue the user is facing.

[1192] 4. Feature extraction and preprocessing: Extract important features (e.g., issues at issue, type of judgment, sentence, etc.) from the retrieved cases and convert them into a format suitable for machine learning models (logistic regression, generative AI models, etc.).

[1193] 5. Generation of predicted judgment patterns: Based on the extracted features and emotional state, a machine learning model is used to generate multiple predicted judgment patterns, such as "Order to pay repair costs and terminate the contract," "Order to pay the full amount of unpaid rent," and "Pay 300,000 yen as a settlement and continue the contract."

[1194] 6. Presentation of results: The server selects the predicted judgment pattern that best suits the user's emotional state from the generated patterns and presents it to the user through the smartphone interface.

[1195] Specific prompt examples

[1196] As a concrete example, the system operates based on the following prompt sentence:

[1197] Case details: "While on night patrol, I spotted a suspicious individual. I was attacked by the individual and temporarily hid. The situation was extremely dangerous, so I called the police."

[1198] Emotional state: "Very anxious and confused."

[1199] Based on this prompt, the system generates an appropriate predictive verdict and provides users with appropriate and specific countermeasures, enabling quick and effective responses in the field of security operations.

[1200] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1201] Step 1: User Input Phase

[1202] Subject: User

[1203] Specific Actions: Users input details of specific legal issues and their emotional state through a smartphone interface.

[1204] Input: For example, the text "I spotted a suspicious person while on night patrol. I was attacked by the person and temporarily hid. The situation was very dangerous, so I called the police." and the emotional state "I was very anxious and confused."

[1205] Output: User input data (legal issue details and emotional state).

[1206] Step 2: Send data

[1207] Subject: Device

[1208] Specific operation: The terminal sends the data entered by the user to the server.

[1209] Input: Details of the legal issue and emotional state entered by the user.

[1210] Output: The user's input data received by the server.

[1211] Step 3: Data reception and analysis

[1212] Subject: Server

[1213] Specific operation: The server receives the user's input data sent from the device, analyzes the received data, extracts features using natural language processing techniques (such as TfidfVectorizer), and analyzes the emotional state using TextBlob.

[1214] Input: User input data sent by the device.

[1215] Output: Extracted features and parsed emotional states.

[1216] Step 4: Search the case law database

[1217] Subject: Server

[1218] Specific operation: The server searches for similar past cases based on the legal issue and emotional state, thereby obtaining relevant cases that match the issue the user is facing.

[1219] Input: Extracted features and parsed emotional states.

[1220] Output: Retrieved relevant case law data.

[1221] Step 5: Feature extraction and preprocessing

[1222] Subject: Server

[1223] Specific operation: The server extracts important features (e.g., issues at issue, type of judgment, sentence, etc.) from the retrieved case data and converts them into a format suitable for machine learning models (logistic regression, generative AI models, etc.).

[1224] Input: The relevant case law data retrieved.

[1225] Output: Preprocessed feature data.

[1226] Step 6: Generate predicted sentence patterns

[1227] Subject: Server

[1228] Specific operation: Based on the extracted features and emotional state, the server uses a machine learning model to generate multiple predicted judgment patterns, such as "order to pay repair costs and terminate the contract," "order to pay the full amount of unpaid rent," and "pay 300,000 yen as a settlement and continue the contract."

[1229] Input: Preprocessed feature data and parsed emotional states.

[1230] Output: Multiple predicted verdict patterns.

[1231] Step 7: Select and present results

[1232] Subject: Server

[1233] Specific operation: The server selects the predicted judgment pattern that best suits the user's emotional state from the generated predicted judgment patterns and presents it to the user through the smartphone interface.

[1234] Input: Multiple predicted judgment patterns and analyzed emotional states.

[1235] Output: The predicted verdict pattern displayed to the user.

[1236] 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.

[1237] 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.

[1238] 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.

[1239] [Fourth embodiment]

[1240] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1241] 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.

[1242] 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).

[1243] 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.

[1244] 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.

[1245] 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).

[1246] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1247] 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.

[1248] 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.

[1249] 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.

[1250] 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.

[1251] 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.

[1252] 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."

[1253] The present invention is a system that utilizes a database of past legal precedents to provide predicted decisions on legal issues. The program processing of this system will be explained below in natural language.

[1254] 1. User Input Phase

[1255] The terminal provides the user with an interface for entering details of the legal issue, including the type of case, specific details of the issue (e.g., amount of compensation, breach of contract).

[1256] Through this interface, users enter details of the legal issue they wish to discuss, such as "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovations during the contract period."

[1257] 2. Data reception and retrieval by the server

[1258] The terminal transmits the data entered by the user to the server.

[1259] The server parses the received user-entered data and generates a search query. Parsing involves the process of extracting important keywords and phrases from the textual input data.

[1260] The server quickly searches the case law database based on the generated search query, and retrieves the top N related cases based on the similarity score and search criteria.

[1261] 3. Feature Extraction and Prediction

[1262] The server preprocesses the case data it receives, converting it into a format suitable for input to machine learning algorithms. This preprocessing includes text cleansing (removing unnecessary information and normalizing) and converting it into structured data.

[1263] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, the type of judgment, applicable laws, and sentencing.

[1264] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the feature values. This algorithm learns from past case data as training data and makes highly accurate predictions for input data.

[1265] 4. Presenting the prediction results

[1266] The server extracts and organizes the generated predicted judgment patterns and converts them into a format that is easy for the user to understand. Specifically, it presents specific judgment results such as "order to pay the full amount of unpaid rent," "order to pay repair costs and terminate the contract," and "payment of 300,000 yen as a settlement and continuation of the contract."

[1267] The terminal displays this predicted judgment result to the user, who can refer to it and use it as information for making legal decisions.

[1268] 5. Feedback and Retraining

[1269] The user can then provide feedback on the prediction results, which may include rating the accuracy of the prediction or providing additional information.

[1270] The server receives user feedback and retrains the machine learning model as needed, continuously improving the accuracy of the system.

[1271] As described above, the present invention provides users with an efficient and reliable tool for quickly assessing legal issues and obtaining a predicted verdict. Through specific implementation methods and processes, the present invention can provide assistance to individuals and businesses in effectively resolving legal issues.

[1272] The processing flow will be explained below.

[1273] Step 1: User Input

[1274] The terminal provides the user with an interface for entering details of the legal matter, including fields for the type of case, the issues at stake, and the specific circumstances.

[1275] Through the interface, the user inputs details of the legal issue they wish to discuss, such as breach of lease agreement, unpaid rent of 500,000 yen, and unauthorized renovations during the lease term.

[1276] The terminal transmits the data entered by the user to the server.

[1277] Step 2: Receiving and analyzing data

[1278] The server receives the user input data sent from the terminal.

[1279] The server analyzes the received data and extracts the type of data (breach of contract, claim for compensation, etc.) and related parameters (amount due, contract terms, etc.).

[1280] Step 3: Search the database

[1281] The server generates a search query based on the extracted data, which is used to search for relevant past cases in a case law database.

[1282] The server uses the search query to quickly search through the case database to retrieve cases similar to the user's legal problem, resulting in the top N most relevant cases.

[1283] Step 4: Feature extraction and preprocessing

[1284] The server analyzes the acquired case data and preprocesses it into a format suitable for machine learning algorithms, including text cleansing (removing unnecessary information and normalizing it) and feature extraction.

[1285] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, type of judgment, and sentence.

[1286] Step 5: Generate predicted verdict patterns

[1287] The server inputs the pre-processed and feature-extracted data into a machine learning algorithm. The machine learning model has already been trained based on past case data and generates multiple predictive verdict patterns for new data.

[1288] Predicted judgment patterns include, for example, "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and continuation of the contract."

[1289] Step 6: Presenting the results

[1290] The server organizes the generated predicted judgment patterns and converts them into a format that is easy for the user to understand.

[1291] The server transmits the predicted judgment result to the terminal.

[1292] The terminal displays the predicted verdict results sent from the server to the user, who can use this information to make legal decisions.

[1293] Step 7: Processing feedback

[1294] The user can then input feedback on the presented prediction results, which may include an evaluation of the prediction accuracy or additional detailed information.

[1295] The server receives user feedback and retrains the machine learning model as needed, continuously improving the system's prediction accuracy.

[1296] These are the specific processing steps of the present invention, which allow individuals and businesses to obtain predicted decisions on legal matters in an efficient and reliable manner.

[1297] Example 1

[1298] 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."

[1299] Currently, there is a lack of systems that provide fast and reliable predictive decisions on legal issues. Many legal issues are complex, and researching past cases and making appropriate legal decisions requires a lot of time and expertise. This makes it difficult for individuals and businesses to receive support to effectively resolve legal issues. In addition, existing systems often lack the accuracy of their machine learning algorithms or retraining capabilities, resulting in inaccurate predictions.

[1300] 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.

[1301] In this invention, the server includes: a means for a user to input legal issues; a means for managing a database of past legal precedents; a means for transmitting user-input data to the server; a means for analyzing the received user-input data and generating a search query; a means for searching the legal precedent database and retrieving relevant legal precedents based on the generated search query; a means for preprocessing the legal precedent data and extracting important features using natural language processing technology; a means for generating multiple predicted verdict patterns using a machine learning algorithm based on the extracted features; and a means for presenting the generated predicted verdict patterns to the user. This enables the server to provide quick and accurate predicted verdicts for legal issues input by the user. Furthermore, the accuracy of predictions can be continuously improved by retraining the machine learning model based on user feedback.

[1302] "User" means any person or entity that utilizes the System to enter legal matters and obtain predicted judgments.

[1303] "Legal Issue" means detailed information about the case or matter that a User seeks to resolve.

[1304] "Means for input" refers to an interface that allows a user to enter details of a legal issue in text form.

[1305] "Means for managing a case law database" refers to technologies and systems that store, search, and manage past case law information in a format that can be referenced.

[1306] "Means for transmitting data to a server" refers to the protocols and technologies used to transmit information entered by a user from a terminal to a server.

[1307] "Means for analyzing data" refers to technology used to process received user-entered data and extract required keywords and phrases.

[1308] "Means for generating a search query" refers to a technology for generating a query for searching a case law database based on the analyzed data.

[1309] The "means for searching a case law database" refers to a technology for quickly searching a case law database using a generated query to obtain relevant cases.

[1310] "Means for pre-processing data" refers to the process of preparing acquired case law data in a format suitable for input into machine learning algorithms.

[1311] "Means for extracting features" refers to technology that extracts important elements from case law data and organizes them into a format that can be used in machine learning models.

[1312] "Means for generating predicted verdict patterns" refers to the process of creating multiple predicted verdict patterns using a machine learning algorithm based on the extracted features.

[1313] The "means for presenting a predicted verdict pattern" refers to a technique for visually presenting the generated predicted verdict pattern to a user.

[1314] "Feedback" refers to the evaluation or additional information provided by the user in response to the predicted results.

[1315] "Retraining" refers to the process of retraining a machine learning model based on feedback from users.

[1316] MODE FOR CARRYING OUT THE INVENTION

[1317] The present invention is a system for providing predicted decisions on legal issues by utilizing a database of past legal cases. The system includes a means for a user to input legal issues, a means for managing the database of legal cases, a means for transmitting and receiving data, a means for analyzing data, a means for generating search queries, a means for searching legal case data, and a means for generating and presenting predicted decisions.

[1318] Hardware and Software Use

[1319] Terminal

[1320] The terminal provides an interface for users to enter legal information, for example, through a web browser or a dedicated application, and includes input fields where users can enter information using a keyboard or touchscreen.

[1321] server

[1322] The server receives the data entered by the user and parses it, using Python and NLP libraries (e.g., NLTK and spaCy) to parse the data and extract important keywords and phrases.

[1323] Based on the received data, a search query is generated for the case law database, and the server executes the generated query using a search engine such as Elasticsearch to retrieve relevant case law data.

[1324] The case data is preprocessed (using libraries such as Pandas and NumPy) and key features are extracted using natural language processing techniques (e.g., NLP models such as BERT).

[1325] Based on the extracted features, multiple predictive verdict patterns are generated using machine learning algorithms (using Scikit-learn and TensorFlow).

[1326] Prediction results

[1327] The server organizes the generated predicted verdict patterns and presents them in a user-friendly format. For example, the results are returned to the terminal in JSON format and rendered in HTML format on the terminal. The user can then view the results using a web browser or similar.

[1328] Feedback and model retraining

[1329] Users can provide feedback on the provided predictions, including a rating of the prediction's accuracy and additional information.

[1330] The server receives feedback and retrains the machine learning model as needed. The retraining process continuously improves the system's predictive accuracy.

[1331] Specific examples

[1332] For example, if a user writes "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," the server will analyze this information and search for similar past cases. The server extracts important features from the case data and applies a machine learning algorithm to present a predicted judgment, such as the following:

[1333] Order to pay all unpaid rent

[1334] "Order to pay repair costs and contract termination"

[1335] "Payment of 300,000 yen as settlement money and continuation of contract"

[1336] Example prompts for generative AI models

[1337] "Generate a predicted verdict based on past precedents for the following legal issue: a breach of lease agreement involving unpaid rent of 500,000 yen and unauthorized renovations during the lease term."

[1338] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1339] Step 1:

[1340] User Input Phase

[1341] The user uses the device interface to enter details of the legal issue, such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," into text input fields provided through a web browser or dedicated application.

[1342] Input: User-entered legal issue details text

[1343] Output: Structured data format (e.g. JSON) sent to the device

[1344] What happens: A web browser renders an HTML form, a user enters information using the keyboard, and clicks the submit button.

[1345] Step 2:

[1346] Data transmission

[1347] The device sends the legal details entered by the user to the server using the HTTPS protocol, encrypting the data.

[1348] Input: User-entered legal issue details

[1349] Output: Encrypted data packet sent to the server

[1350] Specific operation: The device generates an HTTP POST request and sends the data to the server, where the Web API receives the request.

[1351] Step 3:

[1352] Data Analysis and Query Generation

[1353] The server analyzes the received user-entered data, extracts key keywords and phrases, and generates a search query against the case law database.

[1354] Input: Legal issue details received from the device

[1355] Output: Search query against case law database

[1356] What it does: It uses Python and NLP libraries (e.g., NLTK or spaCy) on the server to extract important keywords and generate search queries based on those keywords.

[1357] Step 4:

[1358] Case Law Database Search

[1359] The server searches the case law database based on the generated search query to retrieve relevant cases, calculates a similarity score, and selects the most relevant case.

[1360] Input: search query

[1361] Output: Relevant case law data

[1362] What happens: The server uses a search engine such as Elasticsearch to issue a query and search the database.

[1363] Step 5:

[1364] Data Preprocessing

[1365] The server preprocesses the case data it receives, converting it into a format suitable for input to machine learning algorithms. Preprocessing includes text cleansing and normalization.

[1366] Input: Retrieved case data

[1367] Output: Preprocessed data format

[1368] Specific operation: Using libraries such as Pandas and NumPy on the server side, unnecessary information is removed and the data is normalized.

[1369] Step 6:

[1370] Feature extraction

[1371] The server uses natural language processing techniques to extract important features from the case data.

[1372] Input: Preprocessed case data

[1373] Output: Feature data

[1374] How it works: An NLP model (e.g., BERT) on the server analyzes the text data and extracts important features such as the issues at stake and the type of judgment.

[1375] Step 7:

[1376] Predictive Verdict Generation

[1377] The server generates multiple predicted verdict patterns using a machine learning algorithm based on the extracted features.

[1378] Input: Feature data

[1379] Output: Predicted verdict pattern

[1380] Specific operation: The server runs machine learning models using Scikit-learn and TensorFlow to generate predicted verdict patterns.

[1381] Step 8:

[1382] Prediction results

[1383] The server organizes the generated predicted verdict patterns and sends them to the terminal for presentation to the user. The results are displayed in a format that is easy for the user to understand.

[1384] Input: Predicted sentence pattern

[1385] Output: The resulting data that is presented to the user.

[1386] Specific operation: The result data is sent to the terminal in JSON format, converted to HTML format on the terminal, and displayed. The user can check the results using a web browser or similar.

[1387] Step 9:

[1388] Gathering feedback

[1389] The user can then provide feedback on the provided predictions, including a rating on the accuracy of the predictions and additional information.

[1390] Input: User feedback data

[1391] Output: Feedback data sent to the server

[1392] Specific behavior: The user enters feedback information using a web form on the device and clicks the submit button.

[1393] Step 10:

[1394] Retraining the Model

[1395] The server receives feedback from users and retrains the machine learning model as needed, improving prediction accuracy.

[1396] Input: Feedback data

[1397] Output: Retrained machine learning model

[1398] How it works: The server inputs the feedback data into the machine learning algorithm and retrains the model. The retraining process is performed periodically as a batch process.

[1399] (Application example 1)

[1400] 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."

[1401] Systems that provide predicted decisions for legal issues provide important information by utilizing databases of past legal precedents, but they have difficulty accurately assessing legal risks related to security risks. Another challenge is continually improving the accuracy of predicted decisions based on information entered by users. Furthermore, it is necessary to continuously improve accuracy by incorporating re-learning that reflects user feedback into the system. There is a need to solve these issues.

[1402] 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.

[1403] In this invention, the server includes means for a user to input a legal problem, means for managing a database of past legal cases, means for searching for similar legal cases based on the legal problem, means for extracting specific features from the search results, means for generating a plurality of predicted verdict patterns based on the extracted features, means for presenting the generated predicted verdict patterns to the user, means for inputting details of security risks, and means for performing legal risk assessment, thereby enabling users to predict verdicts for legal problems and security risks with high accuracy.

[1404] Definitions of important terms contained in the claims

[1405] "User" means any person or entity that utilizes the System to input legal issues and security risks and receive predicted adjudications and legal risk assessments.

[1406] A "legal issue" is a legal issue, dispute, or controversy that seeks a legal solution to a specific matter.

[1407] A "case law database" is a database that collects and organizes data on past trials, legal decisions, and verdicts, and is used to resolve similar legal issues.

[1408] "Similar cases" are past cases that have content very similar to the legal issue entered by the user.

[1409] "Features" refer to important elements and parameters extracted from the case law database, and serve as the basic data for predicting verdicts.

[1410] A "predicted judgment pattern" is a set of multiple judgment outcomes generated by a machine learning algorithm based on past case data for a given legal issue.

[1411] "Security risks" refer to threats and dangers to the security of companies and individuals, such as information leaks, system hacking, and internal crimes.

[1412] "Legal risk assessment" is the process of predicting and evaluating the legal consequences and impacts of security risks.

[1413] "Feedback" refers to opinions and suggestions for improvement regarding the use of the system provided by users, which are reflected in improving the accuracy of the system.

[1414] "Retraining" is the process of updating an existing machine learning model to improve its accuracy and performance based on user feedback and additional information.

[1415] MODE FOR CARRYING OUT THE INVENTION

[1416] In carrying out the present invention, a system for assessing legal issues and security risks is constructed, and a specific embodiment thereof is shown below.

[1417] First, the user uses a terminal to input details of legal issues and security risks. This input is done through a dedicated interface. For example, for legal issues, users can enter information such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovations during the contract period," while for security risks, users can enter details such as "Information leak," "System hacking," and "Employee internal crime."

[1418] The entered data is received and analyzed by the server. The server extracts important keywords and phrases from the text data and generates a search query based on these. The server manages a database of past legal precedents and uses the generated search query to quickly search for relevant legal precedents.

[1419] Next, the server extracts specific features from the search results. Natural language processing (NLP) technology is used to extract these features. Preprocessing is then performed, including text cleansing (removing unnecessary information and normalizing it) and conversion to structured data. Then, a machine learning algorithm is used to analyze the data based on the features, and multiple predictive verdict patterns are generated.

[1420] The generated predicted judgment patterns are presented to the user. Specific judgment outcomes, such as "order to pay all outstanding rent," "order to pay repair costs and terminate the contract," or "pay 300,000 yen as a settlement and continue the contract," are displayed as prediction results. Furthermore, a legal risk assessment of security risks is also presented, enabling companies and individuals to take appropriate countermeasures.

[1421] Users can input feedback on these prediction results on their devices, and the server retrains the machine learning model based on the received feedback. This retraining process continuously improves the accuracy of the system.

[1422] The specific implementation of this system uses the following hardware and software: a cloud server (e.g., AWS EC2, Google Cloud), a web framework (Flask), a machine learning library (Sklearn), and a library for saving and loading models (joblib).

[1423] For example, if a user inputs "a past case of a breach of a rental contract resulting in unpaid rent of 500,000 yen," the server will search the database of past legal precedents to extract similar cases. It will then present predicted judgment patterns such as "an order to pay the full amount of unpaid rent" or "an order to pay repair costs and terminate the contract," allowing the user to quickly obtain a legal decision.

[1424] Example prompt sentence:

[1425] "Unpaid rent of 500,000 yen, unauthorized renovations during the rental contract"

[1426] In this manner, the system of the present invention provides users with predicted verdicts and assessments of legal issues and security risks, providing legal decision support.

[1427] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1428] Processing steps of the system that realizes the application example

[1429] Step 1:

[1430] The user enters details about the legal issue or security risk.

[1431] The user enters the specific type of case and details of the problem (e.g., "Violation of rental contract" or "Unpaid rent: 500,000 yen") into the input field on the terminal.

[1432] Step 2:

[1433] The terminal transmits the input data to the server.

[1434] The terminal captures data through the input interface and transmits the data to the server using a secure communication protocol (such as HTTPS).

[1435] Step 3:

[1436] The server analyzes the received data and generates a search query.

[1437] The server uses natural language processing (NLP) techniques to extract important keywords and phrases from the text data, which then generate the search query.

[1438] Step 4:

[1439] The server searches a database of past legal precedents using the generated search query.

[1440] The server uses a high-speed search engine to search the database to identify similar cases, and retrieves the top N relevant cases based on the search criteria.

[1441] Step 5:

[1442] The server extracts specific features from the acquired case data.

[1443] The server preprocesses the data into a format suitable for machine learning algorithms, including text normalization and removal of unnecessary information, and extracts key features (e.g., issues at issue, type of decision, applicable law, sentencing, etc.).

[1444] Step 6:

[1445] The server generates a predicted verdict pattern based on the extracted features.

[1446] The server uses a machine learning model to generate multiple predictive judgment patterns using the extracted features as input. The model is a generative AI model that has learned from past case data.

[1447] Step 7:

[1448] The server presents the generated predicted sentence pattern to the user.

[1449] The server converts the prediction results into a user-friendly format and sends them to the terminal, where the user can check the predicted verdict results on the terminal interface.

[1450] Step 8:

[1451] The user inputs feedback on the provided prediction results.

[1452] The user inputs feedback about the accuracy of the prediction results and additional information into the terminal.

[1453] Step 9:

[1454] The server receives user feedback and retrains the system.

[1455] The server analyzes the feedback data and creates a retraining dataset, which improves the accuracy of the machine learning model.

[1456] In this way, the system provides users with predictive adjudications on legal issues and security risks, and continually improves its accuracy.

[1457] 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.

[1458] The present invention is a system that provides predicted decisions on legal issues by combining a database of past legal precedents with an emotion engine that identifies user emotions. The program processing of this system is explained below in natural language.

[1459] 1. User Input Phase

[1460] The device provides a user with an interface for inputting details of the legal issue and emotional state. The input fields include the type of case, specific details of the issue (e.g., amount of compensation, breach of contract), etc. It also includes sensors and question-type items for inputting or detecting the user's emotional state.

[1461] Through this interface, users can enter details of the legal issue they wish to discuss and, if necessary, their emotional state. For example, they can enter information such as "Violation of rental contract," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period," as well as their own emotional state (e.g., anger, anxiety).

[1462] The terminal transmits the data entered by the user and the detected emotional state to the server.

[1463] 2. Data reception and analysis by the server

[1464] The server receives the user's input data and emotion data transmitted from the terminal.

[1465] The server analyzes the received data and extracts the type of data (e.g., breach of contract, claim for compensation, etc.) and related parameters (e.g., outstanding amount, contract terms, etc.). It also analyzes the user's emotional state using an emotion engine.

[1466] 3. Database Search

[1467] The server generates a search query based on the analysis results, which is used to search for relevant past cases in a case law database.

[1468] The server uses the search query to rapidly search the case database to retrieve cases similar to the user's legal problem, resulting in the top N most relevant cases.

[1469] 4. Feature Extraction and Preprocessing

[1470] The server preprocesses the case data and converts it into a format suitable for machine learning algorithms. Preprocessing includes text cleansing (removing unnecessary information and normalizing) and feature extraction.

[1471] The server uses natural language processing (NLP) technology to extract important features from case data, such as the issues at issue, type of judgment, and sentence.

[1472] 5. Generating predicted verdict patterns

[1473] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the pre-processed and feature-extracted data and the user's emotional state. The machine learning model has already been trained based on past case data and can make highly accurate predictions for new data.

[1474] Predicted judgment patterns include, for example, "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and continuation of the contract."

[1475] 6. Presentation of results

[1476] The server selects the predicted judgment pattern that best suits the user's emotional state based on the emotion engine, thereby providing information in a form that is most acceptable to the user.

[1477] The server transmits the optimal predicted judgment pattern to the terminal.

[1478] The terminal displays the predicted verdict results sent from the server to the user, who can use this information to make legal decisions.

[1479] 7. Feedback and Retraining

[1480] The user can then provide feedback on the provided predictions, which may include a rating on the accuracy of the predictions or additional details.

[1481] The server receives user feedback and retrains the machine learning model as needed, continuously improving the system's prediction accuracy. This retraining also incorporates feedback based on the user's emotional state.

[1482] As a result, the present invention not only provides a predictive verdict for a user's legal problem, but also takes into account the user's emotional state to provide more appropriate advice in a fast and reliable manner. Specific implementation methods and processes are provided to help individuals and businesses effectively resolve legal issues.

[1483] The processing flow will be explained below.

[1484] Step 1: User Input

[1485] The terminal provides a user with an interface for entering details of the legal issue and emotional state, including fields for the type of case, details of the issue (e.g., amount of compensation, breach of contract).

[1486] Through the interface, users enter details of the legal issue they wish to discuss, and also enter their emotional state (e.g., anger, anxiety, etc.) in a field reflecting the emotional state.

[1487] The device sends the details of the legal issue and the emotional state entered by the user to the server in one batch.

[1488] Step 2: Receiving and analyzing data

[1489] The server receives the user's input data and emotion data transmitted from the terminal.

[1490] The server analyzes the received data and extracts the type of input data (e.g., breach of contract, claim for compensation, etc.) and related parameters (e.g., unpaid amount, contract terms, etc.). It also uses an emotion engine to analyze the user's emotional state.

[1491] Step 3: Search the database

[1492] The server generates a search query based on the analysis results, which includes specific keywords related to the law and emotional state.

[1493] The server uses the search query to quickly search the case law database, extracting cases similar to the user's legal problem, and retrieves the top N most relevant cases as search results.

[1494] Step 4: Feature extraction and preprocessing

[1495] The server analyzes the acquired case data and performs preprocessing, which includes text cleansing (removing unnecessary information and normalizing) and feature extraction.

[1496] The server uses natural language processing (NLP) technology to extract important features from case data, including the issues at issue, the type of judgment, and the sentence.

[1497] Step 5: Generate predicted verdict patterns

[1498] The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the results of preprocessing and feature extraction and the user's emotional state. The algorithm is trained based on past case data and can make highly accurate predictions for new data.

[1499] Predicted judgment patterns are generated in specific forms, such as "an order to pay the full amount of unpaid rent," "an order to pay repair costs and terminate the contract," and "a settlement of 300,000 yen and the contract continue."

[1500] Step 6: Presenting the results

[1501] The server selects a predictive judgment pattern that is appropriate for the user's emotional state based on the emotion engine, and this selection provides information in a form that is most acceptable to the user.

[1502] The server transmits the optimal predicted judgment pattern to the terminal.

[1503] The terminal displays the predicted judgment results sent from the server to the user, who then makes legal decisions based on the displayed information.

[1504] Step 7: Processing feedback

[1505] The user can then input feedback on the presented prediction results, which may include an evaluation of the prediction accuracy or additional detailed information.

[1506] The server receives user feedback and retrains the machine learning model as needed, including feedback based on the user's emotional state, to continuously improve the system's prediction accuracy.

[1507] Through these specific processing steps, the present invention can provide a predicted judgment for the user's legal problem and further provide an optimal solution by taking the user's feelings into consideration.

[1508] Example 2

[1509] 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."

[1510] Conventional systems for legal issues are unable to provide predicted decisions that take into account the user's emotional state, making it difficult to provide optimal legal advice quickly. Furthermore, if the accuracy of the predicted decisions provided is low, there is no mechanism to improve the system's performance by incorporating user feedback, which limits the system's usefulness.

[1511] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for a user to input details of a legal issue and their emotional state, means for analyzing received data based on the legal issue and their emotional state, means for managing a database of past legal cases and searching for similar legal cases, means for extracting specific features from the search results and performing preprocessing, means for using a machine learning algorithm to generate multiple predicted verdict patterns based on the extracted features and the user's emotional state, means for selecting and presenting the generated predicted verdict patterns in an optimal form based on the user's emotional state, and means for collecting user feedback and performing re-learning to continuously improve the accuracy of the system. This makes it possible to quickly provide highly accurate predicted verdicts while taking the user's emotional state into consideration.

[1512] "Legal issue details" refers to the specific legal matter the user is attempting to address, including the type of matter and specific details of the issue (e.g., amount of compensation, details of contract breach).

[1513] "Emotional state" refers to the user's current mental or emotional state, and is information that expresses emotions such as "anger," "anxiety," and "sadness."

[1514] "Means for analysis" refers to the technical means for carrying out the process of analyzing input data and extracting data types and important parameters.

[1515] A "case law database" is a database system that systematically accumulates past legal decisions and court cases and manages them in a searchable format.

[1516] A "searching means" is a technical means for efficiently locating information in a database based on specific criteria.

[1517] "Features" are important attributes or metrics extracted from a dataset that machine learning models use to make predictions or classifications.

[1518] "Preprocessing" refers to processes such as data cleansing and normalization that transform data into a form suitable for analytics and machine learning algorithms.

[1519] A "machine learning algorithm" is a mathematical and statistical method for learning from data and making predictions or classifications.

[1520] A "predicted judgment pattern" is a pattern that indicates possible future court outcomes, generated based on analyzed data.

[1521] "Feedback" refers to user-provided evaluations and additional information about the system's predictive accuracy and usability.

[1522] "Retraining" is the process of updating a machine learning model and improving its performance based on new data and feedback collected.

[1523] The present invention is a system that provides predicted decisions for users' legal issues. The system combines a database of past legal cases with an emotion engine that identifies user emotions to provide more reliable legal advice.

[1524] The main components of the system are:

[1525] 1. User Interface - The terminal provides the user with an interface for entering details of the legal issue and emotional state. This interface includes fields for entering the case type, specific details of the issue, and emotional state.

[1526] 2. Data Reception and Analysis - The server receives the user's input data and emotion data sent from the device. The received data is analyzed using analysis software to extract the data type and important parameters. The emotion engine is also used to analyze the user's emotional state.

[1527] 3. Case database search - The server generates a search query based on the received data and analysis results, and searches the case database for the most relevant top N cases.

[1528] 4. Preprocessing and Feature Extraction - The server preprocesses the extracted case data, performing text cleansing and normalization, and then uses natural language processing (NLP) techniques to extract features such as the issues at issue, type of verdict, and sentence.

[1529] 5. Generation of predicted verdict patterns - The server uses a machine learning algorithm to generate multiple predicted verdict patterns based on the pre-processed data and the user's emotional state. This machine learning algorithm is trained based on past case data, allowing for highly accurate predictions.

[1530] 6. Presentation of results - The server selects the predicted verdict pattern that best suits the user's emotional state and sends it to the terminal. The terminal displays the received predicted verdict results to the user, allowing the user to make legal decisions based on this information.

[1531] 7. Feedback collection and retraining - The user inputs feedback on the provided predicted verdict results into the terminal. The server receives this feedback and retrains the machine learning model to improve the system's prediction accuracy.

[1532] Specific examples

[1533] For example, suppose a user wants to review a "violation of a rental contract." The user enters the following information into their terminal: "Unpaid rent: 500,000 yen," "Unauthorized renovation during the contract period," and selects "anger" as their emotional state. This information is sent to the server, where it is analyzed and searched through a precedent database. A predicted judgment pattern is generated based on the user's input and emotional state and presented to the user in the form of, for example, "Order to pay the full amount of unpaid rent," "Order to pay repair costs and terminate the contract," or "Pay 300,000 yen as a settlement and continue the contract."

[1534] Prompt Sentence Examples

[1535] An example of a prompt to input to a generative AI model is as follows:

[1536] text

[1537] User Input: Rental Agreement Violation

[1538] Problem details: Unpaid rent: 500,000 yen, unauthorized renovations during the contract period

[1539] Emotional state: Anger

[1540] System output: Judgment pattern 1: Order to pay the full amount of unpaid rent. Judgment pattern 2: Order to pay repair costs and terminate the contract. Judgment pattern 3: Payment of 300,000 yen as a settlement and continuation of the contract.

[1541] This prompt shows how the system will predict the outcome based on the information entered by the user. As can be seen from this example, the system takes into account the user's emotional state to provide the optimal predicted verdict, enabling it to provide fast and reliable legal advice.

[1542] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1543] Step 1: User Input Phase

[1544] The user inputs details of the legal issue and their emotional state into the terminal. Specifically, they input information such as "Violation of rental agreement," "Unpaid rent: 500,000 yen," and "Unauthorized renovation during the contract period" through the user interface. They also select "Anger" as their emotional state. The input fields are equipped with text boxes and drop-down menus. This input data is temporarily stored in the terminal's memory and prepared for the next processing step.

[1545] Input: Details of the legal issue and emotional state entered by the user into the device

[1546] Output: User input data sent from the device to the server

[1547] Step 2: Data reception and analysis

[1548] The server receives user data sent from the device. This data is sent to the server in a structured data format such as JSON. The server uses data analysis software to analyze the input data and extract the type of legal issue and important parameters. It also uses an emotion engine to analyze the user's emotional state.

[1549] Input: User input data sent from the device to the server

[1550] Output: Analyzed legal issue type, parameters, and emotional state

[1551] Step 3: Search the database

[1552] The server generates a search query based on the analysis results. The generated query is used to search the case law database. For example, conditions such as "Violation of rental contract" and "Unpaid rent: 500,000 yen" are used as a query. The server sends this query to a database management system (DBMS) and quickly searches for related past cases.

[1553] Input: A search query generated based on the analysis results

[1554] Output: A set of relevant case law data

[1555] Step 4: Feature extraction and preprocessing

[1556] The server preprocesses the case data it receives. This preprocessing includes text cleansing and normalization. It uses natural language processing (NLP) techniques to extract features from the case data, such as the issues at issue, type of judgment, and sentence. This converts the data into a format suitable for machine learning algorithms.

[1557] Input: Searched case data

[1558] Output: Preprocessed and feature extracted data

[1559] Step 5: Generate predicted sentence patterns

[1560] The server uses a machine learning algorithm to generate multiple predictive verdict patterns based on the preprocessed and feature-extracted data and the user's emotional state. Because the machine learning model has already been trained on past case data, it can make highly accurate predictions for new data.

[1561] Input: Preprocessed and feature-extracted data, user emotional state

[1562] Output: A list of predicted verdict patterns

[1563] Step 6: Presenting the results

[1564] The server selects the predicted verdict pattern that best suits the user's emotional state from the generated patterns and sends it to the terminal. The terminal then displays the received predicted verdict results to the user, who can then make legal decisions based on this information.

[1565] Input: List of predicted sentence patterns

[1566] Output: The optimal predicted verdict pattern presented to the user.

[1567] Step 7: Feedback and Retraining

[1568] The user inputs feedback on the provided predicted verdict results into the terminal. The feedback includes an evaluation of the prediction accuracy and additional detailed information. The server receives this feedback and retrains the machine learning model. This retraining process continuously improves the system's prediction accuracy.

[1569] Input: User feedback

[1570] Output: The machine learning model after retraining

[1571] Through the above processing steps, the system can quickly provide highly accurate predictive verdicts that take into account the user's emotional state.

[1572] (Application example 2)

[1573] 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."

[1574] Conventional legal problem-solving systems only refer to past legal precedents for legal problems entered by users, and do not provide predicted verdicts that take the user's emotional state into account. This makes it difficult to provide predicted verdicts that are intuitive and easy for users to accept. Furthermore, while security operations require rapid and appropriate responses, conventional systems lack the functionality to support this.

[1575] 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.

[1576] In this invention, the server includes means for a user to input a legal issue and emotional state, means for managing a database of past legal cases, means for searching for similar legal cases based on the legal issue and emotional state, means for extracting specific features from the search results, means for generating a plurality of predicted verdict patterns based on the extracted features and the emotional state, means for presenting the generated predicted verdict patterns to the user, and means for selecting a predicted verdict pattern that best suits the user's emotional state, thereby making it possible to provide an effective and appropriate predicted verdict that takes into account the user's emotional state.

[1577] A "means for user input of legal issues and emotional state" is a device or software that provides an interface through which a user can input or detect details of a specific legal issue and their emotional state.

[1578] The "means for managing a database of past court cases" is a database system for storing past court case data and for efficiently searching and managing it.

[1579] The "means for searching for similar precedents based on the legal issue and emotional state" refers to an algorithm or technology that searches for relevant past precedents based on the legal issue and emotional state entered by the user.

[1580] "Means for extracting specific features from search results" refers to natural language processing and data analysis techniques for extracting important parameters and characteristics from the searched case law data.

[1581] The "means for generating a plurality of predicted verdict patterns based on the extracted features and the emotional state" refers to a machine learning model or algorithm that generates a plurality of possible predicted verdicts using the extracted data and the user's emotional state.

[1582] The "means for presenting the generated predicted judgment pattern to the user" refers to a screen or notification system that displays the predicted judgment result in an easy-to-understand manner for the user.

[1583] The "means for selecting a predicted verdict pattern that is most suitable for the user's emotional state" refers to an algorithm or system that takes into account the user's current emotional state and selects it to present the most appropriate predicted verdict.

[1584] "Means for improving the accuracy of generated predicted verdict patterns using a generative AI model" refers to technologies and methods that use a pre-trained generative AI model to improve the accuracy of the predicted verdicts provided.

[1585] The "means for receiving the user's feedback and re-training the generative AI model" refers to a method for improving prediction accuracy by collecting feedback provided by users and re-training the generative AI model based on that feedback.

[1586] This invention is a security operations support system that combines an interface for inputting users' legal issues and emotional state with a system for managing, searching, and analyzing past case law data. This system can be installed on smartphones, smart glasses, head-mounted displays, and robots. A specific embodiment of this system is described below.

[1587] Hardware and Software

[1588] Smartphone: Provides an interface for users to enter details of their legal issue and their emotional state.

[1589] Server: Receives and analyzes data, searches the case law database, extracts features, generates predicted judgment patterns, and presents the results to the user.

[1590] Natural Language Processing (NLP) technology: Used to extract specific features from retrieved cases.

[1591] Machine learning models: Used to generate multiple predictive verdict patterns based on extracted features and emotional states, including logistic regression and generative AI models.

[1592] TextBlob: A library used to parse emotional states from user input.

[1593] Database management system: Used to efficiently manage past case data.

[1594] Processing Step Description

[1595] 1. User Input Phase: The user inputs details of the specific legal problem and their emotional state through a smartphone interface. For example, they input the text, "I spotted a suspicious person while on night patrol. I was attacked by the person and temporarily hid. The situation was very dangerous, so I called the police." and their emotional state, "I was very anxious and confused."

[1596] 2. Data reception and analysis by the server: The server receives the data entered by the user and extracts features using natural language processing techniques (such as TfidfVectorizer). At the same time, it analyzes the emotional state using TextBlob.

[1597] 3. Database Search: The server searches similar past cases based on the legal issue and emotional state, thereby obtaining relevant cases that match the issue the user is facing.

[1598] 4. Feature extraction and preprocessing: Extract important features (e.g., issues at issue, type of judgment, sentence, etc.) from the retrieved cases and convert them into a format suitable for machine learning models (logistic regression, generative AI models, etc.).

[1599] 5. Generation of predicted judgment patterns: Based on the extracted features and emotional state, a machine learning model is used to generate multiple predicted judgment patterns, such as "Order to pay repair costs and terminate the contract," "Order to pay the full amount of unpaid rent," and "Pay 300,000 yen as a settlement and continue the contract."

[1600] 6. Presentation of results: The server selects the predicted judgment pattern that best suits the user's emotional state from the generated patterns and presents it to the user through the smartphone interface.

[1601] Specific prompt examples

[1602] As a concrete example, the system operates based on the following prompt sentence:

[1603] Case details: "While on night patrol, I spotted a suspicious individual. I was attacked by the individual and temporarily hid. The situation was extremely dangerous, so I called the police."

[1604] Emotional state: "Very anxious and confused."

[1605] Based on this prompt, the system generates an appropriate predictive verdict and provides users with appropriate and specific countermeasures, enabling quick and effective responses in the field of security operations.

[1606] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1607] Step 1: User Input Phase

[1608] Subject: User

[1609] Specific Actions: Users input details of specific legal issues and their emotional state through a smartphone interface.

[1610] Input: For example, the text "I spotted a suspicious person while on night patrol. I was attacked by the person and temporarily hid. The situation was very dangerous, so I called the police." and the emotional state "I was very anxious and confused."

[1611] Output: User input data (legal issue details and emotional state).

[1612] Step 2: Send data

[1613] Subject: Device

[1614] Specific operation: The terminal sends the data entered by the user to the server.

[1615] Input: Details of the legal issue and emotional state entered by the user.

[1616] Output: The user's input data received by the server.

[1617] Step 3: Data reception and analysis

[1618] Subject: Server

[1619] Specific operation: The server receives the user's input data sent from the device, analyzes the received data, extracts features using natural language processing techniques (such as TfidfVectorizer), and analyzes the emotional state using TextBlob.

[1620] Input: User input data sent by the device.

[1621] Output: Extracted features and parsed emotional states.

[1622] Step 4: Search the case law database

[1623] Subject: Server

[1624] Specific operation: The server searches for similar past cases based on the legal issue and emotional state, thereby obtaining relevant cases that match the issue the user is facing.

[1625] Input: Extracted features and parsed emotional states.

[1626] Output: Retrieved relevant case law data.

[1627] Step 5: Feature extraction and preprocessing

[1628] Subject: Server

[1629] Specific operation: The server extracts important features (e.g., issues at issue, type of judgment, sentence, etc.) from the retrieved case data and converts them into a format suitable for machine learning models (logistic regression, generative AI models, etc.).

[1630] Input: The relevant case law data retrieved.

[1631] Output: Preprocessed feature data.

[1632] Step 6: Generate predicted sentence patterns

[1633] Subject: Server

[1634] Specific operation: Based on the extracted features and emotional state, the server uses a machine learning model to generate multiple predicted judgment patterns, such as "order to pay repair costs and terminate the contract," "order to pay the full amount of unpaid rent," and "pay 300,000 yen as a settlement and continue the contract."

[1635] Input: Preprocessed feature data and parsed emotional states.

[1636] Output: Multiple predicted verdict patterns.

[1637] Step 7: Select and present results

[1638] Subject: Server

[1639] Specific operation: The server selects the predicted judgment pattern that best suits the user's emotional state from the generated predicted judgment patterns and presents it to the user through the smartphone interface.

[1640] Input: Multiple predicted judgment patterns and analyzed emotional states.

[1641] Output: The predicted verdict pattern displayed to the user.

[1642] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1643] 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.

[1644] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1645] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1646] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1647] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1648] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1649] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, motorcycles, and other devices, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1650] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1651] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1652] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1653] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1654] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1655] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1656] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1657] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1658] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1659] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1660] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1661] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1662] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1663] The following is further disclosed regarding the above embodiment.

[1664] (Claim 1)

[1665] means for a user to input a legal issue;

[1666] a means of managing a database of past case law;

[1667] means for searching for similar cases based on said legal problem;

[1668] A means for extracting specific features from the search results;

[1669] means for generating a plurality of predicted verdict patterns based on the extracted feature quantities;

[1670] means for presenting the generated predicted judgment pattern to a user;

[1671] A system including:

[1672] (Claim 2)

[1673] 10. The system of claim 1, further comprising means for using the machine learning algorithm to improve accuracy of the generated predicted sentence patterns.

[1674] (Claim 3)

[1675] The system of claim 1 , further comprising means for receiving the user's feedback and retraining the model.

[1676] "Example 1"

[1677] (Claim 1)

[1678] means for a user to input a legal issue;

[1679] a means of managing a database of past case law;

[1680] means for transmitting data entered by a user to a server;

[1681] means for parsing received user input data and generating a search query;

[1682] A means for searching a case law database based on the generated search query to obtain relevant case law;

[1683] A means for preprocessing case law data and extracting important features using natural language processing technology;

[1684] A means for generating a plurality of predicted verdict patterns using a machine learning algorithm based on the extracted features;

[1685] means for presenting the generated predicted judgment pattern to a user;

[1686] A system including:

[1687] (Claim 2)

[1688] 10. The system of claim 1, further comprising means for employing a machine learning algorithm to improve accuracy of the generated predicted sentence patterns.

[1689] (Claim 3)

[1690] 10. The system of claim 1, further comprising means for receiving user feedback and retraining the machine learning model.

[1691] "Application Example 1"

[1692] Extraction of new technical aspects

[1693] In application examples, security risk prediction and legal risk assessment have been added as new technologies, and the system will be improved through re-learning based on these. Specific prediction results and their reliability will also be included.

[1694] (Claim 1)

[1695] means for a user to input a legal issue;

[1696] a means of managing a database of past case law;

[1697] means for searching for similar cases based on said legal problem;

[1698] A means for extracting specific features from the search results;

[1699] means for generating a plurality of predicted verdict patterns based on the extracted feature quantities;

[1700] means for presenting the generated predicted judgment pattern to a user;

[1701] a means for inputting details of the security risk;

[1702] a means of conducting a legal risk assessment;

[1703] A system including:

[1704] (Claim 2)

[1705] 10. The system of claim 1, further comprising means for using the machine learning algorithm to improve accuracy of the generated predicted sentence patterns.

[1706] (Claim 3)

[1707] 10. The system of claim 1, further comprising means for receiving said user feedback and retraining the system.

[1708] "Example 2: Combining Emotion Engines"

[1709] (Claim 1)

[1710] means for a user to input details of a legal issue and emotional state;

[1711] means for analyzing received data based on said legal issue and emotional state;

[1712] A means of managing a database of past cases and searching for similar cases;

[1713] A means for extracting specific features from the search results and performing preprocessing;

[1714] a means for using a machine learning algorithm to generate a plurality of predicted verdict patterns based on the extracted features and the emotional state;

[1715] a means for selecting and presenting the generated predicted judgment patterns in an optimal form based on the emotional state of the user;

[1716] A means of collecting user feedback and retraining to continually improve the accuracy of the system; and

[1717] A system including:

[1718] (Claim 2)

[1719] The system of claim 1, further comprising means for performing preprocessing and feature extraction using the machine learning algorithm and improving the accuracy of the judgment pattern based on the results.

[1720] (Claim 3)

[1721] 10. The system of claim 1, further comprising means for receiving the user feedback and retraining the machine learning model.

[1722] "Application example 2 when combining emotion engines"

[1723] (Claim 1)

[1724] means for a user to input legal issues and emotional states;

[1725] a means of managing a database of past case law;

[1726] means for searching for similar cases based on said legal issue and emotional state;

[1727] A means for extracting specific features from the search results;

[1728] means for generating a plurality of predicted verdict patterns based on the extracted features and emotional states;

[1729] means for presenting the generated predicted judgment pattern to a user;

[1730] means for selecting a predicted verdict pattern that best suits the user's emotional state;

[1731] A system including:

[1732] (Claim 2)

[1733] 10. The system of claim 1, further comprising means for improving accuracy of the generated predicted sentence patterns using a generative AI model.

[1734] (Claim 3)

[1735] 10. The system of claim 1, further comprising means for receiving the user's feedback and retraining the generative AI model. [Explanation of symbols]

[1736] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for a user to input a legal issue; a means of managing a database of past case law; means for searching for similar cases based on said legal problem; A means for extracting specific features from the search results; means for generating a plurality of predicted verdict patterns based on the extracted feature quantities; means for presenting the generated predicted judgment pattern to a user; A system including:

2. The system of claim 1 , further comprising means for using the machine learning algorithm to improve accuracy of the generated predicted sentence patterns.

3. The system of claim 1 , further comprising means for receiving the user's feedback and retraining the model.

Images