• S. Peruvazhuthi
• K. Praveen

The global mortality rate associated with cardiovascular disease (CVD) continues to rise each year. Timely prediction of the disease has the potential to save millions of lives. Machine Learning (ML) algorithms utilized for these tasks are capable of recognizing subtle patterns and risk factors that might not be evident to medical practitioners. However, existing ML methodologies lack the ability to deliver enhanced accuracy and performance. An ensemble technique, the Weighted Majority Voting Ensemble (WMVE) classifier, is implemented to improve the performance of CVD prediction. The WMVE classifier undergoes training and testing using the "Sathvi" dataset, a composite of the Cleveland, Hungarian, Long Beach, and Switzerland datasets. Pearson's correlation technique was utilized for feature selection to remove highly correlated feature. Employing Multilayer Perceptron (MLP), Extreme Gradient Boosting (XGBoost), CatBoost, and Logistic Regression (LR) as the base classifiers, the WMVE model has achieved an accuracy rate of 88.78% and an Area Under the Receiver Operating Characteristic curve (AUROC) of 0.899, outperforming all base classifiers.

cardiovascular disease, machine learning algorithms, weighted majority voting, ensemble technique, feature selection

"A Weighted Majority Voting Ensemble Framework for Cardiovascular Disease Risk Assessment", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 7, page no.f748-f758, July-2024, Available :http://www.jetir.org/papers/JETIR2407595.pdf

"A Weighted Majority Voting Ensemble Framework for Cardiovascular Disease Risk Assessment", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 7, page no. ppf748-f758, July-2024, Available at : http://www.jetir.org/papers/JETIR2407595.pdf