Hybrid Machine Learning with Zebra Optimization Model for Accurate Calorie Burn Prediction

Accurate prediction of calorie burn is essential for personalized health and fitness monitoring, playing a crucial role in optimizing diet, exercise plans, and overall wellness strategies. Consequently, this work proposes a novel hybrid Machine Learning (ML) framework integrating the strengths of the XGBoost classifier and Support Vector Machine (SVM) classifier, enhanced by Zebra Optimization Algorithm (ZOA). The input data is preprocessing to ensure consistency and reliability. XGBoost efficiently handles large, structured datasets and captures complex relationships through gradient boosting, while SVM enhances performance by capturing non-linear patterns in input features such as age, gender, body mass index, activity type, and intensity. To optimize the hybrid model, ZOA is utilized for hyperparameter tuning. The implementation of the proposed model using Python, ensuring a scalable and efficient solution for real-world applications. Model performance is evaluated using key metrics, achieving an accuracy of 96.15%, MAE of 1.71, MSE of 11.24, RMSE of 3.35, and an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> score of 0.99, demonstrating superior predictive accuracy and minimal execution time in calorie burn estimation.