Clustering Semantic Spaces of Suicide Notes and Newsgroups Articles

Historically, suicide risk assessment has relied on question-&-answer tools. These tools, built on psychometric advances, are widely used because of availability. Yet there is no known tool based on biologic & cognitive evidence. This absence often causes a vexing clinical problem for clinicians who question the value of the result as time passes. This paper describes one experiment in a series to develop a tool that combines biological markers with thought markers, & uses machine learning to compute a real-time index for assessing the likelihood of a repeat attempt in the next six months. Unsupervised machine learning was used to distinguish between actual suicide notes & newsgroups. (17 refs.)