A Comprehensive Study: Mental Health Status Prediction Using Machine Learning

Mental health is a growing concern in modern society, especially in high-pressure work environments. With increasing awareness and the availability of large-scale mental health survey data, predictive modeling using machine learning (ML) techniques has emerged as a powerful tool for early diagnosis and support. This research presents a comprehensive pipeline that includes data preprocessing, feature engineering, model building, evaluation, and visualization to predict whether an individual will seek treatment for mental health issues. Multiple models including logistic regression, decision trees, random forests, k-nearest neighbors (KNN), bagging, boosting, stacking, and neural networks are implemented and evaluated. The stacking classifier showed the best performance with an accuracy of 81.7%. The results highlight the importance of factors like work interference, anonymity, and benefits in influencing mental health treatment-seeking behavior.