EmoSupport: A Comparative Study for the Analysis of Mental Health of Undergraduate Students

Mental health disorders, including anxiety, depression, and stress, profoundly impact individuals' well being and necessitate effective early detection for timely intervention. This research investigates the predictive capabilities of machine learning algorithms in assessing anxiety, depression, and stress levels based on questionnaire derived scores. Utilizing a dataset comprising self reported scores obtained through a tailored questionnaire designed for mental health assessment, we delve into the application of Decision Trees, Naive Bayes, Support Vector Machines (SVM), and Random Forests for prediction. Data pre processing involved comprehensive cleaning, encoding categorical variables, and careful feature selection, ensuring the relevance of features in the predictive models. Each algorithm underwent individual implementation, wherein we scrutinized their performances in predicting mental health conditions. Evaluation metrics such as accuracy, precision, and recall were employed to assess the models' proficiency in predicting anxiety, depression, and stress levels. The findings underscore the potential of machine learning in accurately predicting mental health conditions based on questionnaire responses, offering insights into personalized interventions and early detection systems. This study contributes to advancing the understanding of machine learning applications in mental health assessment, highlighting avenues for impactful interventions in mental health care.