International Journal of Scientific Research and Engineering Development

The increasing use of social media platforms and digital devices has significantly influenced human behavior and mental well-being. Continuous exposure to online content, reduced physical activity, irregular sleep patterns, and increased screen time have contributed to rising stress and anxiety levels among individuals. Existing mental health assessment methods are often subjective, time-consuming, and require clinical intervention, which may not be easily accessible to everyone. This paper proposes a Mental Health Prediction System using Social Media Data and Machine Learning to analyze behavioral and lifestyle factors for predicting an individual’s mental health state. The system employs a Random Forest Classifier to categorize mental health conditions into Healthy, Moderate Stress, and Stressed based on features such as social media usage, screen time, sleep duration, physical activity, and emotional indicators. A Flask-based web application is developed to facilitate dataset uploading, model training, and real-time prediction. The proposed system provides consistent and reliable predictions, supporting early mental health awareness and data-driven assessment.

Vishnu Vardhana M, Dr.K.Banuroopa,"Mental Health State Prediction Using Social Media Data and Machine Learning" International Journal of Scientific Research and Engineering Development, V9(1): Page(1912-1920) Jan-Feb 2026. ISSN: 2581-7175. www.ijsred.com. Published by Scientific and Academic Research Publishing.