Abstract:
The number one killer in the world is heart disease, which claims millions of lives each year. Coronary artery disease, often known as CAD, is one of the numerous conditions that may have an adverse impact on the heart. It is one of the leading causes of death among cardiovascular diseases. The prediction of cardiovascular illness is a serious challenge for the industry of clinical data analysis. Studies have shown that machine learning (ML) may assist with decision making and prediction based on data generated by industries such as healthcare, and these industries create a substantial amount of data. In the present body of research, the use of ML techniques to predict heart illness receives only a limited amount of attention. The goal of this study was to investigate which kinds of machine learning classifiers are the most reliable when it comes to achieving such high levels of diagnostic precision. Evaluation and comparison of the effectiveness and efficiency of a number of supervised machine learning algorithms for the prediction of cardiovascular disease were carried out. After utilizing a number of distinct machine learning approaches, such as the Decision tree (DT), Stochastic Gradient Descent (SGD), Random Forest (RF), Adaboost (Ada), and XGB, as well as logistic regression, I discovered that the accuracy of the Stochastic Gradient Descent (SGD) algorithm is 94.66%, making it the most accurate of all the algorithms. And AdaBoost (Ada) is the model with the lowest accuracy, with 93.07%.
