Improving binary crow search algorithm for feature selection

Abstract The feature selection (FS) process has an essential effect in solving many problems such as prediction, regression, and classification to get the optimal solution. For problems, selecting most relevant features of a dataset leads better accuracy with low training time. In this work, hybrid binary crow search algorithm (BCSA) based quasi-oppositional (QO) method is proposed FS on wrapper mode solve problem. QO was employed tuning value flight length BCSA which controlling ability crows find To evaluate performance method, four benchmark datasets have been used are human intestinal absorption, HDAC8 inhibitory activity (IC50), P-glycoproteins, antimicrobial. Accordingly, experimental results discussed compared against other standard algorithms rate, average number selected features, running proven robustness relied high obtained (84.93–95.92%), G -mean (0.853–0.971%), (4.36–11.8) relatively computational Moreover, investigate effectiveness Friedman test declared that supremacy BCSA-QO very evident CSA by minimum producing highest rate. verify usefulness (BCSA-QO) terms accuracy, small