Evaluating Ensemble Models on Imbalanced Data Sets: A Comparative Study across Varied Minority Class Ratios

The complexities of using ensemble models to rectify class imbalance in healthcare datasets are investigated in this study. Using well-known methods such as Random Forest, Gradient Boosting, AdaBoost, CatBoost, LightGBM, XGBoost, BRF and EE we perform an extensive analysis on data-sets that exhibit different levels of class imbalance. The study uses stratified k-fold cross-validation for robust evaluation and the Synthetic Minority Over-sampling Technique (SMOTE) for data augmentation. ROC curves are shown, and performance indicators like as accuracy, precision, recall, F1-score, and area under the curve (AUC) are examined. The results shed light on the complex effects of ensemble models on unbalanced healthcare data-sets and offer directions for further investigation.