Cardiovascular diseases remain the leading cause of global mortality, accounting for approximately 17.9 million deaths annually. Early detection and accurate risk assessment are critical for reducing these fatalities, yet expert cardiological resources are often limited in primary care settings. This research presents a robust machine learning framework designed for the early prediction of heart disease using clinical biomarkers and demographic data. We evaluate and compare four supervised learning algorithms: K-Nearest Neighbors, Support Vector Machines, Decision Tree, and Random Forest. Our methodology utilizes a dataset of 303 patient records with 13 diagnostic parameters, undergoing a rigorous preprocessing pipeline including z-score standardization and one-hot encoding. Experimental results indicate that the KNN algorithm achieved the highest performance with an accuracy of 87.3% and an AUC of 0.91. This system serves as a scalable decision-support tool for healthcare providers.
