Entity Alignment Model Using Centrality and Uncertainty Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing entity alignment methods for knowledge graph construction are time-consuming and labor-intensive, with low alignment accuracy due to the manual selection and annotation of entities, which limits the effectiveness of the entity alignment model.
Innovation Solution
A method for task execution based on a knowledge graph obtained through entity alignment, which involves obtaining a knowledge graph pair, selecting a target entity, determining its centrality and uncertainty, constructing a sample entity pair, training an entity alignment model, and performing entity alignment to merge knowledge graphs and execute tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If entities are selected and annotated manually for training an entity alignment model, then the training process can be performed, but the method is time-consuming and labour-intensive with low alignment accuracy
Solution Approach 1:
The system automatically selects entities with high training value by computing centrality and uncertainty metrics, eliminating the need for manual entity selection and annotation. The entity alignment model trains itself on optimally selected entities without human intervention, achieving both time efficiency and high alignment accuracy.
Solution Approach 2:
The patent introduces two key parameters - centrality (based on graph structure) and uncertainty (based on alignment probability distribution) - to automatically identify and select entities with high training value. By changing from manual selection to parameter-based automatic selection, the system resolves the contradiction between training quality and time consumption.
2Measurement precision
If manual entity annotation is performed to train an entity alignment model, then the model can be trained, but the alignment accuracy is generally not high
Solution Approach 1:
The system automatically identifies high-value training entities through centrality and uncertainty calculations, making the training process self-sufficient without requiring manual annotation expertise. This automated approach both improves alignment accuracy and simplifies the training process.
Solution Approach 2:
By introducing centrality and uncertainty as selection parameters, the system transforms the training process from manual annotation to automated parameter-driven selection, simultaneously improving alignment accuracy and ease of training.
3Productivity
If all entities are used for training the entity alignment model, then comprehensive coverage is achieved, but the training efficiency is low due to lack of focus on high-value entities
Solution Approach 1:
The system extracts and selects only the most valuable entities for training based on centrality and uncertainty metrics, rather than using all entities. This extraction of high-value entities significantly improves training efficiency while maintaining training quality.
Solution Approach 2:
By introducing centrality and uncertainty parameters, the system enables selective extraction of high-value entities from the complete entity set, transforming the training approach from comprehensive but inefficient to focused and efficient.
Data Source
AI summary
Methods, devices, and computer-readable media of task execution based on a knowledge graph obtained by entity alignment are provided. In one aspect, a method includes: obtaining a knowledge graph pair including a first knowledge graph and a second knowledge graph, selecting a target entity by filtering entities in the first knowledge graph, determining a centrality and an uncertainty of the target entity based on the target entity, an entity corresponding to a neighboring node, and entities in the second knowledge graph, constructing a sample entity pair based on the centrality and the uncertainty of the target entity, training a to-be-trained entity alignment model using the sample entity pair, performing entity alignment on to-be-aligned knowledge graphs according to the trained entity alignment model to obtain a merged knowledge graph, and executing a target task based on the merged knowledge graph.