The WHO estimates that deaths due to heart disease are the number one cause worldwide, accounting for around 30% annually taking an estimated 1.5 crores who die due to this disease. In this study, an extension of KNN algorithm known as E-KNN is used and compares with the results of different machine learning methods such as K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Classification and Regression Trees (CART) (Kumar and Thomas in Int J Recent Technol Eng (IJRTE) 9(1) [1]) in the prediction of heart disease. To improve the efficiency of the proposed system, the most important features are selected using chi-square test. The performance and efficiency of the algorithms are evaluated and compared on the basis of accuracy, recall, precision, and F1 score. The results of the proposed algorithm were more accurate with lesser attributes than all attributes. The performance of E-KNN by using 11 attributes has an accuracy value of 90.10%. It is followed by SVM with 89% accuracy.