Heart Disease Prediction Using a Soft Voting Ensemble of Gradient Boosting Models, RandomForest, and Gaussian Naive Bayes

Heart disease is associated with a high mortality rate because it affects a significant number of people around the world. There is a pressing need for improved diagnostic methods that are both effective and accurate. Techniques from the field of machine learning have been put to extensive use on tabular data from the healthcare sector, where they have proven to be effective in prediction and analysis. To address the issue of the traditional machine learning model’s low accuracy, precision, and recall value, we propose a soft voting meta classifier composed of Catboost, Light-Gradient Boosting Machine, Gaussian Naive Bayes , Random Forest, and XGBoost. The proposed soft voting ensemble outperformed the other models used in this experiment, which was conducted on a fused UCI heart disease and Statlog dataset. The proposed soft voting ensemble model achieved 91.85% accuracy and a 0.9344 Area Under The Curve Score.