Obesity Prediction using Smart Watches: A Comparative Analysis of Random Forest and XGBoost

Chronic conditions and obesity are significant global public health issues. This paper investigates the obesity prediction based on health data from smartwatches and compares the performance of Random Forest (RF) and XGBoost (XGB) algorithms. After preprocessing and tuning, XGBoost surpassed RF with accuracy of 99.98%, while RF achieved 85.46%. The results illustrate the promise of wearable technology and machine learning for early obesity diagnosis and personalized treatment by showing XGBoost's superior ability to identify complex patterns.