Predicting Sleep Quality Using 24-Hour Physical Activity Data from Wearable's: A Fitness Tracking Approach

Sujal Ravindra Salgaonkar, Kishori R. Ghule

Abstract: This study suggests a unique method for estimating sleep quality using physical activity data collected over 24 hours via wearable technology like the Apple Watch. We use machine learning models, namely Random Forest and Extreme Gradient Boosting, to investigate the connections between physiological parameters (e.g., heart rate and activity levels) and sleep patterns. Users can improve their sleep quality by understanding how their everyday activities affect them, using the individualised insights generated. According to our research, wearable technology and predictive analytics can improve general health.

Keywords: Apple Watch, wearable technology, machine learning, physical activity, fitness tracking, and sleep quality prediction