A Road Map for Classification of Heart Disease Using Machine Learning Classifier

Abstract

Heart disease becomes one of the most influential diseases that cause a large number of deaths every year around the world. A report by WHO shows in the year 2016 nearly 17 million people gets died due to heart disease every year. The death rate is increasing rapidly day-by-day and it is estimated by WHO that this death ratio will reach the peak of 75 million by 2030. Despite the availability of modern technology and health care system, prediction and diagnosis of heart disease are still beyond the limitations. Currently, the clinical industries and diagnosis centers have a huge of amount data for the diagnosis of heart disease patients. Machine learning algorithms are more useful to find the hidden patterns, discover knowledge from the dataset, and predict correct outcomes. This research proposed an efficient machine learning-based classifier methodology that outperforms the existing similar methodologies. To evaluate the proposed machine learning classifier, we have taken data from the UCI repository. In this study, we have used ZeroR, bagging, M5, and decision table classifier. The M5 classifier produced a good result compared to other classifiers with 0.2726 mean absolute errors.