Comparative analysis of machine learning models for depression risk prediction

Abstract

Depression, affecting 3.8% of the global population and contributing to over 700,000 suicides annually [1], underscores the importance of early detection for effective intervention. This study compares the performance of several machine learning algorithms—Random Forest(RF), XGBClassifier(XGB), LightGBM, and CatBoost—in predicting depression risk based on lifestyle and demographic factors [1]. The models are evaluated using accuracy, precision, recall, F1-score, and advanced techniques such as ROC curves, and calibration plots. The findings aim to provide insights into the potential of ML for early depression risk detection and enhancing intervention strategies.