基于知识图谱链路预测的西药药物风险知识发现OA

[Purpose/significance]The risk information contained in drug instructions is usually incomplete,and some new adverse reactions can only be discovered in actual clinical use.This paper proposes an information organization and knowledge discovery method for pharmacovigilance,in order to timely and accurately identify missing risk knowl-edge in drug instructions.[Method/process]Drug instructions of 8 152 Western medicines are collected as the research data;On the basis of ontology construction,data annotation,and model training,the UIE model is used to jointly ex-tract entity and relationship triplets from the research data;A new knowledge graph link prediction method Comp-GCN-RotatE,is proposed,and performance comparison experiments are conducted with classical RotatE on multiple datasets;Empirical research is conducted by using the proposed method.[Result/conclusion]Compared to the classical Rotate method,research results show that our method has significant improvements in the three indicators,MRR,Hits@3 and Hits@10,and achieve performance of 58.7％,65.6％and 80.5％respectively on the research data.The method proposed can effectively discover drug risk knowledge that is not mentioned in drug instructions but is actual-ly monitored,providing new ideas for pharmacovigilance in China.