TY - JOUR
AU - Griffin, Darrin J.
AB - Background: Identifying precursors that will aid in thediscovery of individuals who may harm themselves or others has long been a focusof scholarly research. Aim: This work set out to determine ifit is possible to use the legacy tokens of active shooters and notes left fromindividuals who completed suicide to uncover signals that foreshadow theirbehavior. Method: A total of 25 suicide notes and 21 legacytokens were compared with a sample of over 20,000 student writings for apreliminary computer-assisted text analysis to determine what differences can becoded with existing computer software to better identify students who may commitself-harm or harm to others. Results: The results support thattext analysis techniques with the Linguistic Inquiry and Word Count (LIWC) toolare effective for identifying suicidal or homicidal writings as distinct fromeach other and from a variety of student writings in an automated fashion.Conclusion: Findings indicate support for automatedidentification of writings that were associated with harm to self, harm toothers, and various other student writing products. This work begins to uncoverthe viability or larger scale, low cost methods of automatic detection forindividuals suffering from harmful ideation.
TI - Analyzing Language in Suicide Notes and Legacy Tokens
JF - Crisis: The Journal of Crisis Intervention and Suicide Prevention
DO - 10.1027/0227-5910/a000363
DA - 2016-01-19
UR - https://www.deepdyve.com/lp/american-psychological-association/analyzing-language-in-suicide-notes-and-legacy-tokens-mMJmUXGcGZ
SP - 140
EP - 147
VL - 37
IS - 2
DP - DeepDyve
ER -