LEVERAGING MACHINE LEARNING FOR EARLY DETECTION OF MENTAL HEALTH ISSUES AMONG HIGHER EDUCATION STUDENTS

Abstract

Mental health disorders among higher education students have become a pressing global concern, with anxiety, depression, and stress significantly impacting academic performance, social relationships, and overall well-being. Early detection and intervention are critical to mitigating these challenges, yet traditional screening methods often fall short due to limited scalability, accessibility, and sensitivity to early signs of distress. This study explores the potential of machine learning (ML) to address these gaps by developing predictive models for identifying students at risk of mental health issues. The research utilized two datasets: a publicly available dataset from Kaggle and a custom dataset collected through an online survey administered to 212 university students. The survey captured diverse dimensions, including demographic, academic, psychological, and social factors, ensuring a comprehensive understanding of the variables influencing mental health. Multiple ML algorithms, including Decision Trees, Support Vector Machines, and Neural Networks, were applied to analyze the data and identify key predictors of mental health risks. The resulting predictive model demonstrated commendable accuracy, highlighting its potential utility for early intervention in educational settings.