AI-Driven Mental Health Predicition: A Scalable Model for Early Intervension in Higher Education Institution

Since mental illness issues among college students have a great impact on academic achievement and overall health, the purpose of this study is to utilize machine learning techniques to create a predictive model that is both accurate and interpretable to identify students' mental health disorders based on psychological, demographic, and academic characteristics. Apart from symptoms like anxiety, panic attacks, and depression, the study found that "Course" and "Year of Study" are the most predictive factors for students' mental health. The study compared different machine learning models, i.e., Logistic Regression, Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Random Forest, to analyze their predictive power. Among these, Logistic Regression and Random Forest offered the optimal balance between precision and recall in handling imbalanced data, with F1 scores of 0.78 and 0.74, respectively. KNN achieved the highest accuracy (94.29%) but at the expense of failing to detect the minority cases well, while SVM achieved perfect precision (1.0) but zero recall (0.0). By identifying crucial determinants like course of study and year of study, universities can formulate targeted support plans, optimize the deployment of resources, and cultivate a more responsive academic culture. This research adds to machine learning-enabled mental health surveillance systems, providing an applied and evidence-informed approach for augmenting student support service in higher education.