XAI-Enhanced Machine Learning for Obesity Risk Classification: A Stacking Approach With LIME Explanations.

Obesity is a growing global health concern, significantly increasing the risk of various chronic diseases. Early detection and classification of obesity risk are crucial for effective prevention and management. This project presents a machine learning approach for obesity risk classification, employing a stacking method with multiple algorithms: Support Vector Machine (SVM), Gradient Boosting Machine (GBM), and CatBoost. The system classifies individuals into seven distinct obesity risk categories: Normal Weight, Overweight Level I, Overweight Level II, Obesity Type I, Obesity Type II, Obesity Type III, and Insufficient Weight. By combining the strengths of these algorithms in a stacked ensemble model, the system aims to improve classification accuracy and robustness. The models are trained and evaluated using Python, with libraries such as scikit-learn, CatBoost, and other machine learning tools. The project provides a reliable solution for obesity risk classification, contributing to personalized healthcare and early intervention strategies.

Keywords: Obesity Risk Classification, Stacking Ensemble, SVM, GBM, CatBoost, Machine Learning, Healthcare, Predictive Modeling, Data Science, Python.