Our bodies’ circulatory systems revolve around the heart, and heart-related illnesses are among the main reasons of increased death rates. Therefore, to comprehend the underlying causes of these ailments and how to detect them in their earliest stages, a comprehensive and accurate investigation is required. Consequently, the working of ML algorithms may prove to be a useful instrument in the future for a cutting-edge healthcare setting and for delivering predictions at a reasonable cost. Several machine learning approaches, including DT, LR, and SVM, RF, GB, NB, were applied to three datasets. In addition to providing insight into the present algorithm, this paper compares the current machine learning algorithm with earlier research and provides an overview of that earlier work. This research compares the outcomes of three distinct models that were developed using various feature selection strategies, including the PSO method (nature-inspired feature selection) and forward feature selection (Wrapper Method). Random forest consistently achieves the maximum accuracy, reaching as high as 98.53%, across all three datasets and feature selection techniques (P.S.O. and forward selection).