Mental Stress Detection in University Students Using Machine Learning

Abstract

Mental stress has become one of the most common psychological challenges among university students due to academic pressure, examinations, financial concerns, social relationships, career uncertainty, and lifestyle changes. Prolonged stress negatively affects students' academic performance, physical health, emotional well-being, and overall quality of life. Early identification of stress levels enables timely counseling and preventive interventions. This study proposes a machine learning-based mental stress detection system that predicts the stress level of university students using demographic, academic, behavioral, and psychological factors. The collected data undergo preprocessing techniques such as missing value handling, categorical encoding, feature scaling, and data balancing before model training. Various supervised machine learning algorithms, including Decision Tree, Random Forest, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Logistic Regression, and Naïve Bayes, are implemented and evaluated using accuracy, precision, recall, F1-score, and confusion matrix. Experimental results indicate that the Random Forest classifier provides the highest prediction accuracy due to its ability to capture complex relationships among multiple stress-related factors. The proposed system offers a reliable, cost-effective, and efficient approach for identifying students at risk of mental stress, thereby assisting educational institutions and mental health professionals in providing timely support and improving student well-being.

Keywords