Calorie Burn Prediction and Fitness Insights with Machine Learning

Accurate prediction of calorie expenditure during physical activity is crucial for effective weight management and personalized fitness planning. This paper presents a machine learning- driven system that leverages an XGBoost Regressor model to estimate calories burned during exercise based on user demographics (age, gender, height, weight) and exercise-related features (duration, heart rate, body temperature). The proposed platform not only predicts calorie burn but also incorporates Body Mass Index (BMI) calculation to provide users with a holistic understanding of their health status. To further enhance user engagement, the system offers tools for estimating exercise duration required to burn a target number of calories and for projecting the time needed to achieve specific weight loss goals based on caloric deficit principles. Distinct roles are defined for Admins and Users: Admins manage user accounts, health tips, and exercise data, while Users can register, log in, access personalized calorie predictions, view health tips, and explore exercise information. Experimental results demonstrate the model’s predictive accuracy and the platform’s potential to support informed, data-driven decisions for fitness and weight management. This integrated approach empowers users to optimize their exercise routines and track progress, fostering improved health outcomes through personalized recommendations.