Improving the Prediction of Heart Disease Using Ensemble Learning and Feature Selection

Abstract

Heart or cardiovascular disease is main cause of mortality. The main objective of developing the proposed model is to increase the accuracy and reliability of predicting the coronary heart disease. This paper attempts in predicting the risk of heart disease more accurately using the techniques of ensemble learning. Moreover, the techniques of feature selection and hyper parameter tuning has been implemented in this work leading to further increase in accuracy. Among the three ensemble techniques, stacking, majority voting and bagging used in this work, the improvement achieved in prediction accuracies is 2.11%, 7.42% and 0.14% respectively. Majority voting has shown the best results in terms of increase in prediction accuracies with an accuracy of 98.38%. Keywords: Heart Disease, Ensemble Learning, Feature selection, Machine Learning