Legal Case Retrieval via Knowledge Graph Construction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current similar-case search methods in litigation are inaccurate and unprofessional due to the lack of clear correspondence between judgment documents and key facts, leading to inefficient retrieval of relevant case information.
Innovation Solution
A knowledge graph-based case retrieval method that constructs a legal case knowledge graph, performs random-walk sampling, and uses cosine similarity calculation to enhance the relevance and comprehensiveness of case information retrieval, improving retrieval speed and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full text of judgment documents is retrieved based only on relevant fields, then retrieval coverage is comprehensive, but retrieval accuracy is low and professional
Solution Approach 1:
The patent segments judgment documents into structured knowledge graphs with distinct node types (case, fact, law, evidence) and relationship types. This segmentation allows precise retrieval of specific case elements while maintaining the complete contextual structure, resolving the contradiction between retrieval accuracy and information completeness.
Solution Approach 2:
The patent transforms flat text data into multi-dimensional knowledge graphs where cases, facts, laws, and evidence form interconnected structures. This dimensional transformation enables simultaneous access to both specific case details and comprehensive contextual information, improving retrieval accuracy without losing information completeness.
2Adaptability or versatility
If judge knowledge graph is constructed based only on judgment document content, then construction is simple, but reference object is single and retrieval usefulness is limited
Solution Approach 1:
The patent creates a universal legal knowledge graph framework that can incorporate multiple data sources (judgment documents, legal provisions, case databases, external legal resources) into a unified structure. This multi-functional knowledge graph serves diverse retrieval needs while maintaining a consistent construction approach, enhancing retrieval usefulness without proportionally increasing complexity.
Solution Approach 2:
The patent introduces standardized knowledge graph schemas and relationship templates as intermediaries between diverse data sources and the retrieval system. These intermediaries enable integration of multiple reference objects with manageable complexity, allowing the system to incorporate rich external legal resources while maintaining structured organization.
3Measurement precision
If multiple retrieval methods are increased to improve retrieval quality, then retrieval comprehensiveness improves, but system availability decreases due to increased operations required
Solution Approach 1:
The patent merges multiple retrieval methods (keyword search, semantic search, relationship-based search) into a unified knowledge graph query system. Users can access comprehensive retrieval functionality through a single interface that automatically leverages the structured relationships in the knowledge graph, improving retrieval quality while maintaining system availability.
Solution Approach 2:
The patent implements automatic query optimization and result ranking within the knowledge graph system. The system automatically selects and combines appropriate retrieval strategies based on the query type and data structure, eliminating the need for users to manually configure multiple retrieval methods and thereby maintaining high system availability while delivering comprehensive retrieval results.
Data Source
AI summary
This application discloses a knowledge graph-based case retrieval method, device and equipment, and a storage medium. The method includes: constructing a legal case knowledge graph based on text information; performing random-walk sampling on node set data constructed based on the legal case knowledge graph, so as to obtain a plurality of pieces of sequence data; training a model by using a word2vec algorithm based on the plurality of pieces of sequence data, so as to obtain an updated target model; obtaining target text information, and analyzing the target text information by using the target model, so as to construct a to-be-retrieved knowledge graph; retrieving the legal case knowledge graph based on the to-be-retrieved knowledge graph, so as to obtain case information associated with the to-be-retrieved knowledge graph; and obtaining outputted case information based on a first similarity and a second similarity of the case information.