CARDIAC DISEASE PREDICTION USING SMOTE AND MACHINE LEARNING CLASSIFIERS

Abstract

Cardiovascular Disease (CVD) is presently the biggest reason of death globally. Clinical data analytics face a huge hurdle when attempting to predict cardiac disease. Massive amounts of raw data generated by the healthcare business are transformed into meaningful insights using machine learning techniques. The goal is to use of machine learning models that can enhance the predictability of cardiac patient survival. This paper employs eight machine learning classifiers: Decision Tree (DT), Extra Tree (ET), Random Forest (RF), Adaptive Boosting (AdaBoost), Ridge Classifier (RC), Linear Discriminant Analysis (LDA) and Light Gradient Boosting Machine (Light GBM) for prediction of cardiac disease. Synthetic Minority Oversampling Technique (SMOTE) is used to resolve the issue of unbalance dataset. Experiment outcomes demonstrate that SMOTE technique improves the accuracy of the selected classifier's output and Random Forest achieves highest accuracy with 95.12% applying SMOTE in predicting the survival of cardiac illness.