Feature Selection for High-Dimensional Datasets through a Novel Artificial Bee Colony Framework

There are generally many redundant and irrelevant features in high-dimensional datasets, which leads to the decline of classification performance extension execution time. To tackle this problem, feature selection techniques used screen out features. The artificial bee colony (ABC) algorithm is a popular meta-heuristic with high exploration low exploitation capacities. balance between both capacities ABC algorithm, novel framework proposed paper. Specifically, solutions first updated by process employing bees retain original ability, so that can explore solution space extensively. Then, modified updating mechanism an strong ability onlooker phase. Finally, we remove scout phase from framework, not only reduce but also speed up algorithm. In order verify our idea, operators grey wolf optimization (GWO) whale (WOA) introduced into enhance capability bees, named BABCGWO BABCWOA, respectively. It has been found these two algorithms superior four state-of-the-art using 12 terms error rate, size subset speed.