Harnessing Wearable Technology for Predictive Health Scoring: A Comparative Study of ML Models

In recent times, increasing reliance on technology and a lack of physical activity have contributed to a rise in health issues such as heart attacks, obesity, and diabetes. To maintain good health, individuals should follow a proper lifestyle and make conscious efforts to stay healthy. However, visiting hospitals frequently and undergoing medical tests can be time-consuming, which many people tend to avoid. Therefore, there is a need for a system that helps individuals monitor and improve their health conveniently. Wearable devices like smart watches can play a significant role in this regard. These devices help users track their health in real time by displaying physical activity parameters such as step count, heart rate, flights climbed, blood oxygen level, calories burnt, and more. These health parameters enable users to monitor their well-being on a regular basis. In this study, these parameters are considered, and machine learning (ML) algorithms are used to calculate the health status of an individual. ML algorithms such as Support Vector Machine (SVM), Gradient Boosting, and Extreme Gradient Boosting (XGBoost) are applied to predict the health score. The performance of all three algorithms is compared, and among them, the SVM algorithm shows the highest accuracy. This approach provides an efficient way to track and assess an individual's health status using wearable devices.