Detecting Suicide Ideation in the Online Environment: A Survey of Methods and Challenges

Abstract

Suicide is a severe mental health problem, and how to curb this social menace has become an important research topic. The advent of the digital age has paved the way for monitoring people's suicidal risks, and many detection approaches have been developed over the years. This article presents an overview of different methods (e.g., technologies, algorithms, etc.) that have been undertaken to identify online suicide ideation. A four-step workflow in this research area is developed during the summarization phase, that is, data collection, data preprocessing, feature engineering, and machine learning (ML) modeling. The current challenges have also been outlined so as to open future directions for research.