Overview This project addresses the pressing issue of mental health among university students by introducing a predictive model for diagnosing mental illnesses. Leveraging machine learning techniques, including Random Forest and Gradient Boosting, the model achieves an impressive 82% accuracy. Through careful feature selection, 35 key survey questions are identified, streamlining the prediction process.
Key Features Machine learning-based prediction of mental health diagnoses. Utilizes Random Forest and Gradient Boosting algorithms. Achieves 82% accuracy in identifying at-risk university students. Feature selection narrows down the survey to 35 key questions for efficient prediction.