Predicting depression using traditional machine learning models

Abstract

Depression is a common mental health disorder that significantly impacts emotional well-being and daily functioning. Early detection is vital but often limited by subjective and resource-intensive clinical assessments. The growing availability of structured health data presents an opportunity for machine learning (ML) to enable scalable, data-driven prediction tools. While deep learning (DL) models are powerful, their need for large datasets and lack of transparency restrict clinical adoption. This study focuses on traditional ML models, which strike a better balance between performance and interpretability, and explores how feature engineering and smart preprocessing can further improve their real-world utility.