Mining heterogeneous networks with topological features constructed from patient-contributed content for pharmacovigilance
作者：
Highlights：
• In this manuscript, we aim at detecting the adverse drug reactions and the drug-drug interactions for pharamcovigilance.
• We collected the data from online health forums to construct health heterogeneous networks that capture the health entities and their associations.
• We extract the topological features from the heterogeneous network. The topological features are meta-paths with different lengths.
• The experiments show that the heterogeneous network mining produce promising results.
摘要
•In this manuscript, we aim at detecting the adverse drug reactions and the drug-drug interactions for pharamcovigilance.•We collected the data from online health forums to construct health heterogeneous networks that capture the health entities and their associations.•We extract the topological features from the heterogeneous network. The topological features are meta-paths with different lengths.•The experiments show that the heterogeneous network mining produce promising results.
论文关键词：Drug safety,Adverse drug reaction,Drug-drug interactions,Heterogeneous networks,Social network mining
论文评审过程：Received 24 January 2017, Revised 12 July 2018, Accepted 18 July 2018, Available online 6 August 2018, Version of Record 23 August 2018.
论文官网地址：https://doi.org/10.1016/j.artmed.2018.07.002
