A supervised machine learning approaches used in heart diseases prediction

Abstract

The healthcare sector consists of vast amounts of data that is often utilized less. The need for this utilization of unused data arises from the requirement of computational decisions supports in smart hospitals for the early diagnosis of different diseases. Heart illness is one of the diseases responsible for so many deaths worldwide, even more than cancer. Heart disease counting is increasing day by day. It may be reduced by early analysis and medication. Utilization of the latest technology may help in the diagnosis of heart disease timely. This work focuses on the likelihoods of occurring heart diseases in terms of percentage, thereby introducing an innovative methodology for classifying heart by various supervised machine learning techniques such as Naive Bayes, KNN, random forest, SVM, decision trees & ensemble methods to handle multivariate datasets in nature. However, only a few of those techniques will be approved for the generation of consistent, precise as well as biologically interpretable rules. Results will be then interpreted to compare the algorithms used on the basis of the accuracy level acquired by each algorithm.