Machine learning uses variations of methods for disease prediction. The present article is aiming to give a thorough explanation of how random forest, decision tree, liner regression are used in our research, especially when combined and applied for the heart disease prognosis. The outcomes of an experiment comparing the implementation of forecasting techniques on the same dataset. In our research, we independently experimented the dataset with random forest, decision tree, and linear regression. The DTKNN1 is our proposed model where we hybrid two machine learning algorithms for achieving the highest accuracy for heart disease prediction. 303 records and 1025 records from different regions combined together in the DTKNN to get 100% accuracy in the prediction of heart disease.