Machine Learning Algorithms with ROC Curve for Predicting and Diagnosing the Heart Disease

Heart diseases are now becoming the leading cause of mortality in India with a significant risk of both males and females. According to the Indian Heart Association (IHA), four people die of heart diseases every minute in India and the age-groups are mainly between 30 and 50. The one-fourth of heart failure morality occurs to people less than 40. A day in India nine hundred people dies below the age of 30 due to different heart diseases. Therefore, it is imperative to predict the heart diseases at a premature phase with accuracy and speed to secure the millions of people lives. This paper aims to examine and compare the accuracy of four different machine learning algorithms with receiver operating characteristic (ROC) curve for predicting and diagnosing heart disease by the 14 attributes from UCI Cardiac Datasets.