Method, medium and device for extracting causal relationship from medical knowledge graph
By converting triple data of the medical knowledge graph into two-dimensional tabular data, and using Bootstrap sampling layer by layer optimization skeleton algorithm and mountain climbing search strategy, the inefficiency and accuracy of causal relationship discovery in the medical knowledge graph in the existing technology is solved, and more efficient and accurate causal relationship extraction is achieved.
Patent Information
- Application Number
- CN202411310647.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-20
- Publication Date
- 2025-03-25
- Estimated Expiration
- 2044-09-20
AI Technical Summary
Existing causal relationship discovery algorithms are difficult to effectively learn causal relationships on graph structure data of medical knowledge graphs, and are easily affected by noise and sample size, resulting in inaccurate learning.
By converting triple data from the medical knowledge graph into two-dimensional tabular data, the layer-by-layer optimization skeleton algorithm based on Bootstrap sampling, combined with the mountain climbing search strategy, the final directed acyclic graph is learned in a directional manner.
It improves the accuracy of causal extraction, overcomes the inefficiency and accuracy of existing algorithms in high-dimensional data and noise environments, and can guide clinical decision-making more scientifically.