Heart disease is the world’s greatest cause of death, affecting the lives of millions of people annually. The diagnosis and prediction of heart disease are critical medical responsibilities that assist cardiologists in correctly classifying patients and treating them accordingly. To lessen the death rate from cardiovascular disorders, this research creates a model that can accurately forecast these conditions. In this work, 10-fold cross-validation was used to optimize single (Naïve Bayes, Logistic Regression, Support Vector Machines), ensemble (AdaBoostM1, Bagging), and hybrid (Logistic Regression + Random Forest, Logistic Regression + Decision Tree) classifiers of machine learning classification. Using a hybrid (Logistic Regression + Decision Tree) classifier, the best accuracy of 98.24% and area under the ROC curve of 0.991 were achieved. This underlying research leads to the conclusion that the hybrid (Logistic Regression + Decision Tree) classifier, with 10-fold cross-validation, has better results in terms of accuracy when compared to the other algorithms. After tuning the parameters of support vector machines, it achieved utmost accuracy of 99.7% using complexity parameter C = 4 and Pearson VII Universal Kernel.