Uncovering Patterns of Depression in Engineering and Medical Students Using Machine Learning

This paper presents a study on the evaluation of mental health challenges faced by engineering and medical students in Bangladesh, utilizing advanced machine learning and statistical methods. Data were collected through comprehensive questionnaires, focusing on socio-demographic, academic, and mental health variables. The dataset, including responses from 834 students, was preprocessed and split into two subsets: one for medical students and another for engineering students. The study utilized Student's T-Test and Fisher's Exact Test to identify 16 key features associated with mental health challenges. Following feature selection, a Support Vector Machine (SVM) classifier was used to evaluate the impact of these features on depression prediction. The evaluation metrics included accuracy, ROC AUC, precision, recall, and F1 score, allowing for a comprehensive assessment of the models' performance. By leveraging advanced machine learning techniques, this research seeks to uncover patterns and predictors of depression that traditional studies often overlook. The findings aim to provide targeted insights into the factors contributing to mental health challenges in these student populations, guiding effective interventions. The paper emphasizes the global relevance of the study, as mental health challenges among university students are prevalent worldwide.