Random forest prediction of suicide attempts in adolescents reached an AUC of 0.72

n = 2809, data from ages 14-15 and prediction at ages 16-17. The data was from a longitudinal study in Australia.

The authors used several interesting techniques to deal with the unbalanced data, and also used predictor selection to prune unnecessary variables.

---

References §

• Machine learning-based prediction for self-harm and suicide attempts in adolescents