Construction Method of Drug Adverse Reaction Knowledge Graph Based on Knowledge-Driven Technology Path
10.3870/j.issn.1004-0781.2025.04.011
- VernacularTitle:基于知识驱动技术路径的药品不良反应知识图谱构建方法
- Author:
Wei REN
1
;
Yan XU
;
Qing WANG
;
Zitong PEI
;
Tao WANG
;
Ying LIU
;
Yexiang ZHANG
Author Information
1. 清华大学药物警戒信息技术与数据科学创新中心,北京 100084
- Publication Type:Journal Article
- Keywords:
Adverse drug reactions;
Knowledge-driven;
Knowledge graph;
Supervision;
Intelligence
- From:
Herald of Medicine
2025;44(4):565-569
- CountryChina
- Language:Chinese
-
Abstract:
Objective In view of the potential threat of adverse drug reaction(ADR)to patients'health and the limitations of traditional monitoring methods,this study investigates the construction method of an ADR knowledge graph based on a knowledge-driven technology path to promote the intelligent upgrading of pharmacovigilance field.Methods In this study,an adverse drug reaction knowledge base was established,and the data within the knowledge base were preprocessed.The BERT-CRF model was utilized to identify and extract the relevant entities,relationships and attribute information,and knowledge alignment was carried out by combining the rule method and statistical method.The knowledge fusion data was stored in the Neo4j graph database to construct the knowledge graph.Results Based on the established knowledge base,this study explored the method of constructing knowledge graph of adverse drug reaction by knowledge-driven technology,laying the groundwork for data mining and the development of intelligent auxiliary system tools based on knowledge graph.Conclusions The study on the construction method of an ADR knowledge graph based on a knowledge-driven technology path provides a new perspective and technical path for the intelligent upgrading of pharmacovigilance field.It holds important theoretical and practical significance for improving drug safety,protecting patients'health,and promoting the healthy development of pharmaceutical industry.
