Suicide Note Classification Using Natural Language Processing: A Content Analysis

Abstract

Suicide is the second leading cause of death among 25-34 year olds and the third leading cause of death among 15-25 year olds in the United States. In the Emergency Department, where suicidal patients often present, estimating the risk of repeated attempts is generally left to clinical judgment. This paper presents our second attempt to determine the role of computational algorithms in u nderstanding a suicidal patient's thoughts, as represented by suicide notes. We focus on developing methods of natural language p rocessing that distinguish between genuine and elicited suicide notes. We hypothesize that machine learning algorithms can categorize suicide notes as well as mental health professionals and psychiatric physician trainees do. The data used are comprised of suicide notes from 33 suicide completers and matched to 33 elicited notes from healthy control group members. Eleven mental health profess ionals and 31 psychiatric trainees were asked to decide if a note was genuine or elicited. Their decisions were compared to nine different machine-learning algorithms. The results indicate that trainees accurately classified notes 49% of the time, mental health professionals accurately classified notes 63% of the time, and the best machine learning algorithm accurately classified the notes 78% of the time. This is an important step in developing an evidence-based predictor of repeated suicide attempts because it shows that natural language processing can aid in distinguishing between classes of suicidal notes.

Genuine and simulated suicide notes: An analysis of content

“…Giles [19] suggests that the genre is intuitively accessible to everyone who wishes to produce a suicide note for whatever reason and recognised three genuine suicide note types: relational and health disputes, mental and physical health and intentional escape. In line with psychological phenomenology assertions, the discrepancies between genuine and simulated suicide notes may be affected by the fact that the experience of suicidal individuals cannot be directly observable by and thence accessible to an outsider [21]. A fictitious note writer, in order to gain an insight into the suicidal individual's cognition, would need to assume an insider's perspective [22].…”

“…More specifically, the stereotypical image of the suicidal mind as unsound or broken may suggest that a product of such a mind must be internally flawed too [34]. Another explanation may be found in the area of psychological phenomenology whose main assumption is that subjective experience cannot be grasped by an onlooker [21][22][23] -hence the discrepancies between genuine and simulated suicide notes.…”

Proceedings of the Fourth Workshop on Computational Linguistics and Clinical Psychology –- From Linguistic Signal to Clinical Reality

“…Text analysis applied to the study of extraordinary events like murder and suicide has tended to focus mostly on classification (e.g., Pestian et al, 2010), with a mind toward prevention (e.g., Brynielsson et al, 2013). Though such work is certainly interesting and worthwhile, the goals of this project are more oriented toward understanding (in a phenomenological rather than explanatory sense).…”

Natural language processing in mental health applications using non-clinical texts

“…The aim was to explore the difference between why people think others suicide, and why individuals actually do. Pestian et al (2010) extracted from this book a corpus of thirty-three genuine (written by people who completed suicide) and thirty-three fake notes (written by people simulating a suicide note). They showed how difficult they are to distinguish manually: In their experiments, mental health professionals achieved an accuracy of 63%, whereas psychiatry trainees were only 49% accurate (i.e.…”