Published Paper Details:

To detect and predict heart attack, we use the dataset from the UCI repository by the use of machine learning classifiers. Enhancing accuracy, facilitating early detection, and allocating healthcare resources as efficiently as possible are the objectives. Our main goal is to identify most suitable classifier which we can use for diagnostic applications. To predict heart disease, we use several machine learning approaches are used and compare their accuracy to get best result.This study discovered that the RF approach obtained 78% accuracy coupled with 100% sensitivity and specificity utilizing a heart attack dataset that we gathered from Kaggle three-classification based on k-nearest neighbor (KNN), decision tree (DT), and random forests (RF) algorithms. To construct a model training and testing of data evaluated. Accuracy, precision, recall, and F1-score were used to evaluate the models. With an accuracy of 80.33%, the SVM model outperformed the others.