Mental Health Prediction Using Artificial Intelligence

Mental health problems such as stress, anxiety, and depression are increasing worldwide. Early detection is important, but many people avoid professional help due to stigma or lack of access. This project proposes an AI-based Mental Health Prediction System using machine-learning techniques and symptom-based inputs. A structured dataset containing self-reported symptoms was used. Logistic Regression, Support Vector Machine (SVM), and Random Forest algorithms were trained and evaluated. Among these, the Random Forest classifier achieved the highest accuracy ranging between 90%. The system classifies users into four categories: Normal, Stress, Anxiety, and Depression. A Streamlit web application was developed to allow users to enter symptom ratings and receive instant predictions. Feature analysis showed that sleep issues, anxiety level, and fatigue were the most influential factors. The system provides fast predictions within two seconds. Although not a replacement for clinical diagnosis, this system supports early mental-health screening and encourages timely professional assistance.