A Data Preprocessing and Stacking Ensemble Learning Model for Improved CHD Prediction

Coronary Heart Disease (CHD) is a significant public health concern, posing a substantial burden on individuals, families and healthcare systems, necessitating accurate prediction and timely diagnosis for effective management. This research paper presents a data-driven approach for enhancing CHD prediction using meticulous preprocessing to address null values, ensuring data integrity, dataset resampling to address imbalance and Min-Max normalization which further enhances the comparability of features for the Framingham CHD dataset. A Stacking Ensemble Model with Random Forest as the meta classifier is proposed, that combines the predictive capabilities of the base classifiers including Logistic Regression, KNN, SVM, Decision Tree and XGBoost. The proposed approach yields exceptional results with an accuracy of 97.39