Breast Cancer Diagnosis Approach Based on Meta-Heuristic Optimization Algorithm Inspired by the Bubble-Net Hunting Strategy of Whales

Abstract

This paper proposes a novel meta-heuristic optimization algorithm, called Whale Optimization Algorithm (WOA) to select optimal feature subset for classification purposes of Wisconsin Breast Cancer Database (WBCD). WOA is considered one of the recent bio-inspired optimization algorithms presented in 2016. A set of measurements are used to evaluate the different algorithm over WBCD from the UCI repository. These measurements are precision, accuracy, recall and f-measure. The obtained results are analyzed and compared with those from other algorithms published in breast cancer diagnosis. The experimental results show that WOA algorithm is very competitive for breast cancer diagnosis. Also it has been compared with seven well known features selection algorithms; genetic algorithm (GA), principle component analysis (PCA), mutual information (MI), statistical dependency (SD), random subset feature selection (RSFS), sequential floating forward selection (SFFS) and Sequential Forward Selection (SFS). It obtains overall 98.77 % accuracy, 99.15 % precision, 98.64 % recall and 98.9 % f-score.