Heart Disease Prediction Using Bayesian Optimized Classification Algorithms

Heart disease is a pressing global public health issue, and accurate prediction of cardiovascular diseases is crucial for early intervention and prevention. This paper introduces a methodology for predicting heart disease using Bayesian-optimized classification algorithms. To select important features, a Genetic Algorithm (GA) is employed. The study investigates the effectiveness of popular machine learning algorithms such as Random Forest (RF), Support Vector Machines (SVM), Adaptive Boosting (AdaBoost), and extreme gradient boosting (XGB). These algorithms are fine- tuned using Bayesian optimization to optimize their hyperparameters. The research utilizes an online dataset from the UCI repository, which includes clinical factors used to train and evaluate predictive models. The results demonstrate that incorporating Bayesian optimization significantly enhances the performance of these classification algorithms, resulting in improved prediction accuracy. Among them, RF achieves the highest accuracy rate at 86.87%, followed by SVM with 83.75%, AdaBoost with 85.00%, and XGB with 84.37%.