A Novel Feature Selection Algorithm using Multi-Objective Improved Honey Badger Algorithm and Strength Pareto Evolutionary Algorithm-II

An important task for classification is feature selection that removes the redundant or irrelevant features from dataset. Multi-objective approach mainly proposed by many researchers. However, these approaches failed to maintain higher accuracy while removing redundancy in features. In this work, a wrapper based technique with hybrid of Multi Objective Honey Badger Algorithm (MO-HBA) and Strength Pareto Evolutionary Algorithm-II balance between removal redundancy. Classification improvement are considered as multi-objective optimization functions technique. The Levy flight algorithm utilized initialize population enhance ability exploration exploitation MO-HBA. regularized Extreme Learning Machine used classify selected To evaluate performance technique, eighteen benchmark datasets results compared four well known techniques terms accuracy, hamming loss, ranking mean value, standard deviation, length features, training time. achieved maximum 100% value 80. minimum deviation 0.0092, 0.0003, 0.018 0.001 respectively. experimental show can give improved large scale datasets.