Applied Unsupervised Learning for Pattern Recognition of Depression cases within a Young Adult Population

Abstract

This study uses unsupervised learning to identify depression patterns in Mexican university students. By analyzing demographic, academic, and psychological factors, it aims to uncoversub groups with similar depression profiles and identify risk and protective factors. The study compares clustering algorithms and evaluates their performance using metrics like the Silhouette Coefficient and Davies-Bouldin Index. This research contributes to the field of machine learning in mental health and may improve support services for students at risk of depression.