Integrating Survey Data and Biotechnological Biomarkers for Student Stress Prediction Using Machine Learning

Stress among students is a growing concern, affecting both mental and physical health and potentially leading to chronic conditions if unaddressed [1], [2]. An accurate assessment of stress levels is essential for timely intervention and effective mental health support. Traditional methods rely on self-reported surveys and clinical evaluations, which may be subjective and resource intensive [3], [4]. In this study, we present a machine learning-based approach to predict stress levels (High, Medium, Low) using survey data collected from students. Three models, Logistic Regression, Random Forest, and Support Vector Machine (SVM), were trained and evaluated. The SVM model achieved the highest accuracy of 91.38. The model performance was assessed using precision, recall, F1-score, and confusion matrices. Additionally, we reviewed physiological biomarkers, including cortisol, serotonin, heart rate variability, and sleep cycles, to provide a biotechnological perspective on the effects of chronic stress. This study demonstrates the feasibility of integrating survey data and biological insights for the early detection and intervention of student stress.