Knowledge Graph Generation via Target Element Merging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing knowledge graph technologies face challenges in efficiently generating and mining knowledge graphs, particularly in handling dynamic relations and large-scale data integration.
Innovation Solution
The method involves obtaining an initial knowledge graph, determining target map elements meeting preset property information requirements, and merging these elements to generate a knowledge graph. This process includes merging nodes and edges based on specific property information criteria, reducing graph complexity, and improving data integration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If traditional knowledge graph generation methods are used to handle large-scale data integration, then comprehensive knowledge coverage is achieved, but computational complexity and processing time increase significantly
Solution Approach 1:
The patent segments the knowledge graph generation process into distinct modules: entity recognition module, relation extraction module, and graph construction module. This segmentation allows each module to process specific tasks independently, reducing overall computational complexity while maintaining comprehensive data integration capabilities
Solution Approach 2:
The patent performs preliminary data preprocessing and entity recognition before relation extraction and graph construction. By preparing data in advance and organizing it into structured formats, the system reduces the computational burden during the main processing stages, enabling efficient handling of large-scale data
2Measurement precision
If detailed property information requirements are applied during knowledge graph generation, then mining precision is improved, but processing time increases
Solution Approach 1:
The patent applies different property information requirements to different parts of the knowledge graph based on local needs. Critical entities and relations receive detailed property extraction, while less important elements use simplified processing, thereby maintaining high mining precision for key information while reducing overall processing time
Solution Approach 2:
The system dynamically adjusts the depth of property information extraction based on the importance and type of entities being processed. For high-priority entities, comprehensive property extraction is performed, while for lower-priority entities, only essential properties are extracted, optimizing the balance between precision and processing time
3Device complexity
If the knowledge graph structure is simplified to reduce complexity, then processing efficiency improves, but information completeness may be compromised
Solution Approach 1:
The patent extracts only the most essential and frequently used entities and relations to form a simplified core knowledge graph structure. Less critical information is stored in an extended database that can be queried when needed, thereby reducing graph complexity while preserving information completeness through on-demand access
4Loss of information
If comprehensive entity and relation extraction is performed, then knowledge coverage is improved, but the scale of the knowledge graph increases leading to higher computational burden
Solution Approach 1:
The knowledge graph is segmented into core entities, extended entities, and metadata layers. This segmentation allows the system to maintain comprehensive knowledge coverage across all layers while enabling efficient processing by operating primarily on the compact core layer and accessing extended layers only when necessary
Data Source
AI summary
The present disclosure provides a method for generating a knowledge graph, a method for mining relation and an apparatus for generating a knowledge graph. The method for generating a knowledge graph includes: obtaining an initial knowledge graph, wherein the initial knowledge graph comprises a plurality of nodes having connection relationships; determining a plurality of target map elements meeting preset property information requirements from the initial knowledge graph, in which the target map elements comprise target nodes, or target edges; and merging the plurality of the target map elements to generate a knowledge graph.