Reading Between the Signs: Predicting Future Suicidal Ideation from Adolescent Social Media Texts
Blum, P.; Liscio, E.; Zhang, R.; Figueroa, C.; Murukannaiah, P. K.
Show abstract
Suicide is a leading cause of death among adolescents (aged 12-18), yet predicting it remains a significant challenge. Many cases go undetected because young people often do not contact mental health services. In contrast, young people often share their thoughts and struggles online in real time. To utilize this communication channel, we propose a novel task and method: predicting suicidal ideation and behavior (SIB) from online forums before an adolescent explicitly expresses suicidal ideation on a forum. This predictive framing, where selfdisclosure is not used as input at any stage, is largely unexplored in the suicide prediction literature. We introduce EO_SCPLOWARLYC_SCPLOW-SIB, a transformer-based model that sequentially processes the posts a user writes and engages with to predict whether they will write a SIB post. Our model achieves a balanced accuracy of 0.73 in predicting future SIB on a Dutch youth forum, demonstrating that such tools can offer a meaningful addition to traditional methods.
Matching journals
The top 8 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Machine Learning Models Predict the Emergence of Depression in Argentinean College Students during Periods of COVID-19 Quarantine 92%
- Deep Multimodal Representations and Classification of First-Episode Psychosis via Live Face Processing 91%
- Passive sensing data predicts stress in university students: A supervised machine learning method for digital phenotyping 91%
Similar papers in this journal
- Developing an automatic system for classifying chatter about health services from Twitter: A case study for Medicaid 92%
- Tracking private WhatsApp discourse about COVID-19: A longitudinal infodemiology study in Singapore 92%
- Information retrieval in an infodemic: the case of COVID-19 publications 91%
Similar papers in this journal
Similar papers in this journal
- Building a Best-in-Class De-identification Tool for Electronic Medical Records Through Ensemble Learning 93%
- Inferring global-scale temporal latent topics from news reports to predict public health interventions for COVID-19 91%
- RiskPath : Explainable deep learning for multistep biomedical prediction in longitudinal data 90%
Similar papers in this journal
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.