基于临床数据与多语料验证低噪知识识别模型及建立方法
By using a low-noise knowledge recognition model based on clinical data and multiple corpora, combined with knowledge graphs and expert verification, the problems of high noise and poor accuracy in medical knowledge graphs are solved, thereby improving the support capability and accuracy of medical insurance risk control.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 上海金仕达卫宁软件科技有限公司
- Filing Date
- 2021-12-03
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for constructing medical knowledge graphs suffer from problems such as insufficient manually labeled data, high noise levels, and poor accuracy. This results in insufficient accuracy and interpretability of the knowledge in medical knowledge graphs, making them unable to effectively support medical insurance risk control.
We employ a low-noise knowledge recognition model based on clinical data and multiple corpora. By integrating multiple corpora and expert verification, we construct a low-noise knowledge recognition model, use knowledge graphs for preliminary recognition, combine clinical data for weight allocation and training, construct an accuracy matrix, and integrate prediction models to identify knowledge relationships.
It improves the accuracy and reliability of medical knowledge graphs, enhancing their support for medical insurance risk control, particularly in the accuracy and efficiency of unsupported diagnosis and prediction of diagnostic resource consumption risks.
Smart Images
Figure CN114358956B_ABST