Breast cancer is one of the most common diseases among women. Early and accurate diagnosis is essential to improve treatment outcomes and patient survival. In this study, we propose two hybrid classification models for breast cancer diagnosis by combining Twin Support Vector Machine (Twin SVM) with two metaheuristic optimization algorithms: Starfish Optimization Algorithm (SFOA) and Particle Swarm Optimization (PSO). These models, named SFOA-Twin SVM and PSO-Twin SVM, use a Radial Basis Function (RBF) kernel and aim to optimize the dual variables in the Twin SVM framework. Both approaches are applied to improve classification accuracy. A 30-run cross-validation experiment is conducted on a breast cancer dataset to evaluate their performance. The results show that the PSOTwin SVM model achieves higher accuracy and better overall performance compared to the SFOA-Twin SVM model.

Keywords