Accurate Prediction of Heart Disease Using Machine Learning: A Case Study on the Cleveland Dataset

Abstract

Heart disease remains one of the leading causes of mortality worldwide, with diagnosis and treatment presenting significant challenges, particularly in developing nations. These challenges stem from the scarcity of effective diagnostic tools, a lack of qualified medical personnel, and other factors that hinder good patient prognosis and treatment. The rise in cardiac disorders, despite their preventability, is primarily due to inadequate preventive measures and a shortage of skilled medical providers. In this study, we propose a novel approach to enhance the accuracy of cardiovascular disease prediction by identifying critical features using advanced machine learning techniques. Utilizing the Cleveland Heart Disease dataset, we explore various feature combinations and implement multiple well-known classification strategies. By integrating a Voting Classifier ensemble, which combines Logistic Regression, Gradient Boosting, and Support Vector Machine (SVM) models, we create a robust prediction model for heart disease. This hybrid approach achieves a remarkable accuracy level of 97.9%, significantly improving the precision of cardiovascular disease prediction and offering a valuable tool for early diagnosis and treatment.