Event Detection for Suicide Understanding

Abstract

Suicide is a serious problem in every society. Understanding life events of a potential patient is essential for successful suicide-risk assessment and prevention. In this work, we focus on the Event Detection (ED) task to identify event trigger words of suicide-related events in public posts of discussion forums. In particular, we introduce a new dataset for ED (called SuicideED) that features seven suicidal event types to comprehensively capture suicide actions and ideation, and general risk and protective factors. Our experiments with current state-of-the-art ED systems suggest that there is still room for improvement of ED models in this domain. We will publicly release SuicideED to support future research in this important area.

Tasks

Reproductions