Heart Disease Prediction: Hyper Parameter Tuning Based Ensemble Learning and Voting Classifiers Approach

As per the observations of World Health Organization (WHO), heart related issues are the prominent reason for the loss of life and is becoming more common every year. Timely therapy raises the possibility of cure. For this, early prediction and good diagnosis are essential. Today's technology development allows for more accurate and precise predictions. Researchers in prediction modeling are interested in machine learning (ML) methods because of their accuracy, precision, and predictability. An attempt is made here to predict cardiac diseases using ML techniques, including Logistic Regression (LR), Decision Tree (DT), Support Vector Machine (SVM), Random Forest (RF), and Multi-Layer Perceptron(MLP). The cardiac disease dataset was acquired from the Kaggle website, and the prediction was done using an ensemble of the standard ML methods based on hyper parameter tuning and voting classifiers. Finally, the findings are compared and concluded based on the ensemble model voting approach and the predictive performance metrics.