Prediction of Cardio Vascular Diseases Using Calibrated Machine Learning Model

Cardio Vascular Disease (CVD) uncovers various conditions that influence Heart Attack and cause more death rates in the recent decades. It is the need of hour to get efficient, reliable approaches to predict the possibilities of CVDs and timely management of the treatment to save lives from disease. Several methods have been used by researchers to predict CVDs. The aim of this research paper is to predict the Cardio Vascular Disease with Calibrated Classification Model. This model detected the likelihood of CVDs more accurately and effectively. The experimental results depict that the Gaussian NaïveBayes (GNB) Model with Sigmoid Calibration achieved highest accuracy score with 90.21%, when compared with Logistic Regression, SVM, Decision Tree, Random Forest, K-nearest Neighbor Algorithms and less Brier Score losses than isotonic calibration.