一种基于大模型的自然资源图谱补全方法
By combining the structural information of a large language model and a natural resource knowledge graph, and employing knowledge hints and structural prefix adapters, the problem of difficulty in completing long-tail entities in the natural resource knowledge graph is solved, improving completion efficiency and accuracy, and expanding application prospects.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SUN YAT SEN UNIV
- Filing Date
- 2024-11-18
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies face difficulties in completing long-tail entities in natural resource knowledge graphs. The fine-tuning process is time-consuming and complex, resulting in poor knowledge graph completion effects.
A natural resource graph completion method based on a large model is adopted, which combines the reasoning ability of a large language model with the structured information of the natural resource knowledge graph. Through knowledge hints and structural prefix adapter processing, the completion ability of long-tail entities is improved and the consumption of fine-tuning resources is reduced.
It significantly improves the ability to complete long-tail entities and the reasoning accuracy of complex natural resource relationships, increases the efficiency of knowledge graph completion, enhances the adaptability to diverse natural resource scenarios, and reduces fine-tuning costs.
Smart Images
Figure CN119647571B_ABST