Machine Learning Models for Early Heart Disease Prediction: A Comparative Analysis

One of the many major health issues, heart disease is one that primarily impacts the cardiovascular system, encompassing the heart and its associated organs. Individuals who smoke, suffer from high blood pressure, have elevated cholesterol levels, maintain poor dietary habits, lack physical activity, or are overweight or obese face an increased risk of developing heart disease. The most frequently diagnosed cardiovascular condition is coronary artery disease (CAD), while other types include congestive heart failure, arrhythmias, and congenital heart defects. This condition is often referred to as cardiovascular disease. Prompt and accurate diagnosis is essential for saving lives and preventing additional complications. Recent advancements in machine learning diagnostic techniques have demonstrated enhanced reliability and effectiveness in detecting heart disease. In this study, we applied three machine learning models: K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Naïve Bayes (NB) and Decision Tree, to assess their performance in predicting heart disease using the UCI dataset. The algorithms take into account various factors such as age, gender, resting blood pressure, cholesterol levels, blood sugar levels, and ECG results. The results reveal that the KNN model achieves an accuracy of 87%, the SVM model attains an accuracy of 87%, NB 79% and the Decision Tree model surpasses the others with an impressive accuracy of 96%.