Comparative Study of Supervised Machine Learning Methods for Prediction of Heart Disease

Abstract

With the ever-growing medical data, it became possible to use the data for prediction of diseases using Machine Learning (ML) methods. ML methods have been widely employed in healthcare. In the study, some of the mostly used ML methods such as Support Vector Machine, Naïve Bayes classifier, Random Forest, Decision tree, and K-Nearest Neighbor were used for prediction of heart disease. Further, we aim to provide a comparative analysis of the ML algorithms applied for heart disease prediction using their accuracy metrics. Dataset for the study was taken from Kaggle in .csv format, where data mining steps such as data collection, data cleaning, data preprocessing, and exploratory data analysis have been done. The study highlights the ML methods used for classification, providing the comparative analysis between them. As a result, it was concluded that the random forest gave the highest accuracy rate with the dataset used in the study.