Cardiovascular Disease Diagnosis Prediction System Using Machine

Cardiovascular disease is a significant cause of global mortality. Every day, more people suffer this terrible disease. Thus, a complete heart illness classification system is essential. This research used numerous machine learning methods to classify this deadly disease. The Cleveland dataset is collected from the UCI repository and then processed to get the desired form. We then utilized SMOTE to control unbalanced data. After-prepossessing 10-Stratified Fold Cross-Validation separated the dataset. This research uses machine learning to classify and predict cardiovascular disease. Inspired by our findings, we developed an ML-based CVD ensemble forecasting approach. The proposed ensemble approaches include K-Nearest Neighbor (KNN), Extra Trees, Decision Tree, Random Forest, Gradient Boosting, and Ada Boost classifiers. Voting-based ensemble classifiers train many base models or estimators and average their results to forecast. This voting classifier uses the Hard and Soft Voting Classifiers. The hard Voting ensemble approach produces the best outcomes with 87.5% accuracy.