Opposition-based learning tunicate swarm algorithm for biomedical classification

Disease classification is one of the most difficult tasks when developing medical diagnosis software. Swarm Intelligence (SI) algorithms are used in the diagnosis and treatment of the disease. By combining ML classification algorithms and SI algorithms, we can improve the speed, performance, reliability, and accuracy of diagnosing a particular disease on the current system. This paper introduces an improved version of the Tunicate Swarm Algorithm (TSA) hybridized with the Opposition-Based Learning (OBL) strategy called TSA-OBL for classification approach based on the k-Nearest Neighbor (kNN) classifier. For experimental purposes, we have employed four datasets, namely, Breast Cancer Wisconsin (Diagnostic), Leukemia2, Base Brain T91, and Lung Cancer. The results help to verify the good performance of the proposal, which can find the best solutions with high accuracy in comparison with other methodologies and improve the original algorithm results.