“…The other goal is to put forward a comprehensive comparative study of some variants of the ABC, PSO and GA algorithms on wrapper feature selection in terms of the classification performance and the feature subset size for the future studies of researchers. To establish the second goal, seven algorithms, which are binary PSO (BPSO) [22], new velocity based binary PSO (NBPSO) [23], quantum inspired binary PSO (QBPSO) [24], discrete ABC (DisABC) [21], angle modulated ABC (AMABC) [25], modification rate based ABC (MRABC) [26] and genetic algorithms (GA) [27] are employed, and 10 benchmark datasets, including various classes, instances and features are chosen from the UCI machine learning repository [28]. To our knowledge, the employed algorithms except for BPSO and GA are the first time to be used in feature selection, and a comprehensive comparative analysis on feature selection is not very common in the literature.…”