Software defect prediction based on enhanced metaheuristic feature selection optimization and a hybrid deep neural network

Kun Zhu, Shi Ying, Nana Zhang, Dandan Zhu

Abstract: Highlights•Build a feature selection algorithm (EMWS) for defect prediction using WOA and SA.•Propose a unified defect predictor called WSHCKE by CNN and KELM.•Conduct extensive experiments for feature selection and defect prediction.