Heart failure is a chronic condition characterized by the weakening of the heart muscle and its reduced ability to efficiently pump blood. This can lead to various significant health issues, including breathlessness, fatigue, and swelling in the legs and feet. Heart failure is the primary cause of mortality in the United States, emphasizing the criticality of early detection and proper treatment to prevent severe complications. Ensemble learning represents a machine learning technique that enhances prediction accuracy and robustness by combining predictions from multiple models. It has proven effective in various tasks, including predicting the risk of heart failure. This research paper introduces an ensemble learning model specifically designed for heart failure risk prediction. The model combines predictions from three distinct machine learning algorithms: Multi-Layer Perceptron(MLP), random forests (RFs), Sequential Parallel Tree and AdaBoost. Training is performed on a dataset comprising both heart failure and non-heart failure patients, with evaluation conducted on a separate test set. The outcomes demonstrate that the ensemble learning model surpasses the individual machine learning algorithms in terms of performance on the test set. Specifically, the ensemble learning model achieves an impressive accuracy of 94%, a sensitivity of 93%, and a specificity of 95%.