Automatic reasoning case trial finding method

Abstract

The invention discloses an automatic reasoning case trial finding method. The method comprises the following steps: collecting and preprocessing data; legal knowledge element extraction and machine learning; constructing a legal knowledge graph and sorting rules; and designing and realizing an inference rule engine. The invention aims to provide a system for assisting judges and lawyers in case trial and ascertaining. The method is based on big data and legal knowledge graph construction, has automatic reasoning and inference capabilities, and can process and analyze case-related materials, evidences and legal clauses to support decision making in a trial process. According to the invention, the case trial efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the field of inference engines and logical reasoning, and particularly to a method for automatically reasoning about case trial and verification. Background Art

[0002] Machine Learning: Machine learning techniques can be applied to the analysis of case-related data and pattern recognition. Methods such as supervised learning, unsupervised learning, and reinforcement learning can be used to classify cases, predict case outcomes, or discover patterns and regularities in case information.

[0003] Data Mining: Data mining techniques are used to discover useful patterns, relationships, and trends from large datasets. In case trials, data mining can help discover key information related to cases, such as evidence, related events, etc.

[0004] Knowledge Graph: A knowledge graph is a graphical data representation method used to display relationships between entities. In the legal field, a knowledge graph can integrate legal knowledge, case information, legal provisions, etc., providing a basis for automatic reasoning.

[0005] Inference Engine: An inference engine is a software component used to perform reasoning and inference operations. In case trials, an inference engine can perform reasoning based on known legal rules, case facts, etc., assisting judges or lawyers in making decisions.

[0006] Legal Knowledge and Rules: Automatic reasoning methods need to understand and apply this legal knowledge in order to perform reasoning and judgment in case trials. Summary of the Invention

[0007] Aiming at the problems existing in the prior art, the present invention provides a method for automatically reasoning about case trial and verification, which is a tool for assisting decision-making for judges and lawyers, improving the efficiency and accuracy of case trials.

[0008] The object of the present invention is achieved through the following technical solutions.

[0009] A method for automatically reasoning about case trial and verification, the steps including: 1) Collection and preprocessing of data: The collection of the data includes collecting a large amount of case data, legal texts, and relevant materials to establish a dataset; the preprocessing includes cleaning, de-duplicating, and annotating the data. 2) Extraction of legal knowledge elements and machine learning: Using machine learning techniques, analyze legal texts to extract legal knowledge elements; establish a model of legal knowledge elements and train and optimize it. 3) Construction and Rule Sorting of Legal Knowledge Graph: The construction of the legal knowledge graph includes entities, relationships, and attributes; Sort out relevant rules and logics in the legal knowledge graph, establish a rule base, covering common legal rules and clauses in case trials; 4) Design and Implementation of Inference Rule Engine: Design the architecture and logic of the inference rule engine, determine the execution process of inference algorithms and inference rules; implement the core functions of the inference rule engine, including inference and deduction of case information, and derivation of the process of trial ascertainment.

[0010] This step also includes system integration and interface design: Integrate the legal knowledge graph, rule base, and inference rule engine into the system; design a user interface to provide an intuitive operation interface and interactive experience.

[0011] This step also includes testing and evaluation: Conduct functional testing, performance testing, and security evaluation on the system to ensure the system is stable and reliable; conduct testing and evaluation for actual cases to examine the effectiveness and accuracy of the system in actual applications.

[0012] This step also includes continuous improvement and optimization: Collect user feedback and system usage data, analyze problems and improvement spaces, and continuously optimize the performance and functions of the system based on the feedback and data to improve the intelligence and user experience of the system.

[0013] The entities described in step 3) include legal provisions and case subjects, the relationships include application and citation relationships, and the attributes are case attributes, including parties, case numbers, and filing dates.

[0014] The rule classification in the rule base described in step 3) includes: Legal norms: Provisions directly from legal texts; Judicial interpretations: Specific interpretations of legal provisions issued by judicial institutions; Case rules: Legal application rules extracted from specific cases; Procedural rules: Provisions on litigation procedures, including burden of proof and limitation of action; Substantive rules: Provisions on substantive rights and obligations.

[0015] Compared with the prior art, the advantages of the present invention are as follows: The present invention aims to assist judges and lawyers in case trials and ascertainment systems. The present invention is constructed based on big data and legal knowledge graphs, has the ability of automatic inference and deduction, can process and analyze case-related materials, evidence, and legal provisions to support decision-making in the trial process. The present invention improves the efficiency and accuracy of case trials. Description of the Drawings

[0016] Figure 1 The rule result diagram provided by the present invention.

[0017] Figure 2 The rule configuration flowchart provided by the present invention.

[0018] Figure 3 The case flowchart provided by the present invention.

[0019] Figure 4 The rule derivation flowchart provided by the present invention. Detailed implementation manners

[0020] The present invention will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0021] As Figures 1-4 shown, a method for automatically reasoning about case trial and verification includes the following steps: 1. Data collection and preprocessing: Collect a large amount of case data, legal texts and related materials to establish a data set. Perform preprocessing tasks such as data cleaning, deduplication, and annotation on the data to ensure data quality and usability.

[0022] 2. Extraction of legal knowledge elements and machine learning: Use machine learning techniques to analyze legal texts and extract legal knowledge elements such as legal provisions and case precedents. Establish a model of legal knowledge elements and train and optimize it to adapt to different types of legal texts. Use a large amount of case data and legal texts, and combine machine learning algorithms to extract and analyze legal knowledge elements. By identifying, extracting, and generalizing legal knowledge elements, support and a foundation are provided for subsequent rule data splicing.

[0023] 3. Construction of legal knowledge graph and rule sorting: Construct a legal knowledge graph, including entities (such as legal provisions, case subjects), relationships (such as "apply to", "cite", etc.), and attributes (case attributes: parties, case numbers, filing dates, etc.).

[0024] Sort out the relevant rules and logics in the legal knowledge graph, establish a rule library, covering common legal rules and provisions in case trials. Include legal texts, case laws, legal interpretations, etc., and sort out and organize relevant legal rules and provisions. Through the systematic organization and integration of legal knowledge, a rich knowledge base and rule library are provided for the inference rule engine.

[0025] Among them, the common rule classifications are as follows: Legal norms: directly derived from the provisions in legal texts; Judicial interpretation: Specific interpretations of legal provisions issued by institutions such as the Supreme People's Court; Case rules: Rules of law application extracted from specific cases; Procedural rules: Provisions regarding litigation procedures, such as the burden of proof, limitation of action, etc.; Substantive rules: Provisions regarding substantive rights and obligations, such as the validity of contracts in the Contract Law, etc.; 4. Design and implementation of the reasoning rule engine: Design the architecture and logic of the reasoning rule engine, determine the reasoning algorithm and the execution process of the reasoning rules. Implement the core functions of the reasoning rule engine, including the reasoning and inference of case information, and the derivation of the process of examination and determination. The reasoning rule engine utilizes entities, relationships, and rules in the legal knowledge graph, combines with the reasoning algorithm to conduct reasoning and inference of case information. The engine can perform logical reasoning based on case facts, legal provisions, and previous cases, thereby generating conclusions of examination and determination.

[0026] Including: Intelligent element analysis and dynamic rule assembly: This system uniquely introduces an intelligent element analysis mechanism, which can automatically identify and analyze business elements from different data sources with diverse characteristics. Based on these elements, the system can assemble personalized rule chains that conform to the current business scenario in real-time and flexibly, achieving a leap from "one-size-fits-all" to "tailor-made".

[0027] Construction of a multi-dimensional rule library: We have constructed a multi-dimensional and extensible rule library system, covering a wide range of business fields and decision-making dimensions. These rules are not only independently configurable, but also support logical association and priority sorting among each other, ensuring the accuracy and efficient execution of the rule engine when dealing with complex business logics.

[0028] Visual rule configuration interface: To lower the technical threshold of rule configuration, we have developed an intuitive and easy-to-use visual rule configuration interface. Users do not need to have in-depth knowledge of complex programming knowledge, and can quickly construct rule processes that meet business requirements through simple operations such as dragging and connecting lines, greatly improving the usability and popularity of the rule engine.

[0029] Assistance in case trial and determination: The reasoning rule engine derives the examination and determination from the legal knowledge graph and the rule library, providing assistance and support for judges and lawyers in case trial and determination. Judges and lawyers can use the reasoning conclusions and reasoning processes provided by the system to assist in case trial and determination, improving trial efficiency and decision-making accuracy.

[0030] 5. System integration and interface design: Integrate the legal knowledge graph, rule base, and inference rule engine into the system. Design a user interface to provide an intuitive operation interface and interactive experience, facilitating judges and lawyers to use the system for case trial and fact-finding.

[0031] 6. Testing and Evaluation: Conduct functional testing, performance testing, and security evaluation on the system to ensure its stability and reliability. Test and evaluate the system with actual cases to examine the effectiveness and accuracy of the system in practical applications.

[0032] 7. Continuous Improvement and Optimization: Collect user feedback and system usage data, analyze problems and areas for improvement. Based on the feedback and data, continuously optimize the performance and functions of the system to enhance its intelligence and user experience.

[0033] The above steps constitute the specific implementation of the method for automatic inference in case trial and fact-finding. This process requires crossing multiple fields such as data processing, machine learning, knowledge graph construction, and inference engine design.

Claims

1. A method for automatic reasoning case investigation and verification, characterized by the steps of include: 1) Data collection and preprocessing: The data collection includes collecting a large amount of case data, legal texts and related materials to establish a data set; the preprocessing includes cleaning, deduplication and labeling of the data; 2) Extraction of legal knowledge elements and machine learning: Use machine learning technology to analyze legal texts and extract legal knowledge elements; establish a model of legal knowledge elements and train and optimize it; 3) Construction of legal knowledge graph and sorting out of rules: The construction of the legal knowledge graph includes entities, relationships and attributes; Sort out the relevant rules and logic in the legal knowledge graph and establish a rule library covering the common legal rules and clauses in case trials; 4) Design and implementation of inference rule engine: Design the architecture and logic of the inference rule engine, determine the execution process of the inference algorithm and inference rules; implement the core functions of the inference rule engine, including reasoning and inference of case information, and deriving the trial and investigation process.

2. The method for automatic reasoning case investigation and verification according to claim 1 is characterized in that The steps also include system integration and interface design: integrating the legal knowledge graph, rule base and reasoning rule engine into the system; Design user interface to provide intuitive operation interface and interactive experience.

3. The method for automatic reasoning case investigation and verification according to claim 1 is characterized in that The steps also include testing and evaluation: functional testing, performance testing and security evaluation of the system to ensure that the system is stable and reliable; Conduct tests and evaluations on actual cases to verify the effectiveness and accuracy of the system in actual applications.

4. The method for automatic reasoning case investigation and verification according to claim 1 is characterized in that The steps also include continuous improvement and optimization: collecting user feedback and system usage data, analyzing problems and room for improvement, and continuously optimizing system performance and functionality based on feedback and data to improve system intelligence and user experience.

5. The method for automatic reasoning case investigation and verification according to claim 1 is characterized in that The entities in step 3) include legal clauses and case subjects, the relationships include applicable and citation relationships, and the attributes are case attributes, including parties, case numbers, and case filing dates.

6. The method for automatic reasoning case investigation and verification according to claim 1 is characterized in that The rule categories in the rule base described in step 3) include: Legal norms: provisions directly derived from legal texts; Judicial interpretation: a specific interpretation of a legal provision issued by a judicial body; Case rules: legal application rules extracted from specific cases; Procedural rules: provisions regarding litigation procedures, including the burden of proof and the statute of limitations; Substantive rules: provisions regarding substantive rights and obligations.

Images