Optimizing Maternal Mental Health: A Study on Boosting Algorithms for Suicidal Tendencies Prediction in Postpartum Depression

This study comprehends boosting algorithms to forecast suicidal behaviors in postpartum depression patients. The data is collected from a survey on postpartum depression symptoms and used for facilitating the procedure. Five boosting methods are used to enhance performance in accuracy, precision, F-1 score, AUC score, and other metrics. XGB outperforms the other boosting algorithms with an accuracy of 99.15%, while LightGBM and Catboost achieved accuracies of 96.15% and 94.02% respectively. This approach may play a crucial role in the field of mental health by evaluating the risk of postpartum depression and perhaps saving lives.