Performance Comparison of Various Machine Learning Approaches to Identify the Best One in Predicting Heart Disease

Abstract

Nowadays, machine learning is growing fast to be more popular in the world, especially in the healthcare field. Heart diseases are one of the most fatal diseases, and an early prediction of such disease is a vital task for many medical professionals to save their patient’s life. The main contribution of this research is to provide a comparative analysis of different machine learning models to reach the most supporting decision for diagnosing heart disease with better accuracy as compared to existing models. Five models namely, K-Nearest Neighbor (KNN), Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), and Extreme Gradient Boost (XGB), have been introduced for this purpose. Their performance has been tested and compared considering different metrics for precise evaluation. The comparative study has proven that the XGB is the most suitable model due to its superior prediction capability to other models with an accuracy of 91.6% and 100% on two different heart ailments datasets, respectively. Both datasets were acquired from the heart diseases repositories where dataset_1 was taken from the University of California, Irvine (UCI) and dataset_2 was from Kaggle.