"DIGITAL-STRESS" cycle: As daily screen time increases, happiness and sleep quality consistently decrease, while stress levels rise. This high-stress, low-sleep profile was found to be the single most important predictor of a user's "At-Risk" status, which our machine learning models could identify with over 95% accuracy. Crucially, the data also pointed to a clear solution, showing that regular exercise was the strongest factor correlated with both lower stress and better sleep, providing a data-driven strategy to counteract the negative effects of high screen time.