Heart disease has become a common cause of death worldwide in recent years. People’s way of living changes, dietary habits, office working cultures, and other factors have all played a role in this worrisome problem around the world. The best way to stop this disease is to develop a method that will detect early symptoms and hence save more lives. With the help Machine Learning (ML) algorithms, researchers can predict the likelihood of developing cardiovascular disease in people who are at risk. It is critical to develop a precise and dependable technique to have early disease prediction by automating the task and therefore achieving efficient disease management. Several academics have described their efforts to develop the best feasible technique for predicting heart disease in previous publications. The goal of this study is to compare alternative algorithms for predicting cardiac disease. The results of important data mining techniques are presented in this work, which can be utilized to construct a highly efficient and accurate prediction model that will aid doctors in minimizing the number of people killed by heart disease. This study compares the metrics for prediction of heart disease for 6 ML algorithms which are “Logistic Regression” (LR), “Decision Tree” (DT), “Random Forest” (RF), “Support Vector Machine” (SVM), “Gaussian Naïve Bayes” (GNB) and “k-Nearest Neighbor” (kNN).