Machine Learning for Suicide Ideation Detection in Twitter
TEM Journal · 27.02.2026 · 485
DOI: 10.18421/tem151-44 ↗Kurz gefasst
Fachartikel in der Zeitschrift TEM Journal, online erschienen am 27.02.2026. Verfasst von Silvia Rostianingsih und 3 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Twitter is a popular platform for sharing thoughts and emotions, but some tweets indicate suicidal ideation, requiring urgent attention. This research develops a tool to detect such content in Indonesian tweets, categorizing them as suicidal ideation or explicit invitations to suicide. Ten algorithms, including Support Vector Machine, Logistic Regression, Random Forest, and Bidirectional Encoder Representations from Transformers, were tested using Grid search for optimization. Support Vector Machine and Logistic Regression achieved 90% accuracy, with Random Forest and others reaching 89%. Tests under four binary settings showed the best performance in S 1, where five algorithms, including Support Vector Machine and Logistic Regression, achieved 93% accuracy.
Zitieren
APA
Rostianingsih, S., Wijaya, S. C., Budhi, G. S., & Adipranata, R. (2026). Machine Learning for Suicide Ideation Detection in Twitter. TEM Journal, 485. https://doi.org/10.18421/tem151-44
Vancouver
Rostianingsih S, Wijaya SC, Budhi GS, Adipranata R. Machine Learning for Suicide Ideation Detection in Twitter. TEM Journal. 2026:485. doi:10.18421/tem151-44
BibTeX
@article{Rostianingsih2026,
author = {Rostianingsih, Silvia and Wijaya, Steve Christopher and Budhi, Gregorius Satia and Adipranata, Rudy},
title = {Machine Learning for Suicide Ideation Detection in Twitter},
journal = {TEM Journal},
year = {2026},
pages = {485},
doi = {10.18421/tem151-44}
}
EUVIMED