Heart Disease Prediction Using Machine Learning-Based Voting Ensemble

Heart disease continues to be one of the global death drivers, pinpointing the need of detection at an early stage alongside timely and precise diagnosis. In the context of clinical diagnosis, the application of traditional machine learning models is rather restricted due to the low accuracy predictive models. This paper's system proves to be more accurate and precise than the norm ML models relying on the heart disease dataset. The aim of current project is heart disease prediction using hybrid Random Forest-CNN model with ensemble methods such as Voting, XGBoost and Extra Tree Classifier applied. The RFCNN model uses the selection of features by Random Forest and pattern recognition by CNN to grab in data and to complex relationships among the data. The new method ensures stable and robust heart disease diagnosis. Making use of ensemble approach gives an advanced solution and reduces complex dependencies.