Heart Disease Classification Using Support Vector Machines Enhanced with Metaheuristic Optimization Algorithms

Abstract

Heart disease is still one of the major causes of death globally, for which early and accurate diagnosis is of prime importance. This work introduces an efficient approach to classify heart disease using Support Vector Machines (SVMs) with innovative metaheuristic optimization techniques. The method is tested on the Cleveland Heart Disease dataset from the UCI Machine Learning Repository with 303 patients having 14 features. Particle Swarm Optimization (PSO), Grey Wolf Optimizer (GWO), and Whale Optimization Algorithm (WOA) metaheuristic algorithms were used for optimization of SVM parameters. The performance measures used for the comparison are accuracy, precision, recall, and F1-score. The highest classification accuracy (93.7%) is produced by the proposed GWO-SVM, which is relatively better than that of PSO-SVM (91.2%) as well as WOA-SVM (89.6%). The proposed work yields superior predictive performance compared to the contemporary SVM and existing models in the literature. The outcomes of this work emphasize the integration of metaheuristic algorithms with machine learning models for clinical decision support in the diagnosis of heart disease.