Heart attack prediction using machine learning

Abstract

The most prevalent kind of cardiovascular illness is a heart attack, which may or may not have symptoms. The damage to the heart muscle increases with delayed treatment, which increases the risk of mortality. More than 10 million people die each year from heart attacks, and many of them may be avoided if heart attacks could be accurately predicted. To estimate the likelihood of suffering a heart attack, five different machine learning algorithms are used on the Public Health Heart Attack dataset. Several evaluation metrics, including accuracy, recall, precision, ROC curve, and F-score, were used to evaluate the models. All the models-MLP, RBF, SVM, KNN, and RF-achieved significant accuracies of more than 75%, with KNN having the greatest overall performance