Identifying adverse drug event information in clinical notes with distributional semantic representations of context

Aron Henriksson, Maria Kvist, Hercules Dalianis, Martin Duneld

Abstract: Highlights•A corpus of Swedish clinical notes was annotated for adverse drug event information.•Detecting adverse drug events in clinical notes can support pharmacovigilance.•Modeling context with distributional semantics yielded better predictive models.•Distributed word representations allowed more context information to be incorporated.•Inter-sentential relations between drugs and disorders/findings are hard to detect.