ISSN : 2319-7323

INTERNATIONAL JOURNAL OF COMPUTER SCIENCE ENGINEERING

ABSTRACT

Title	:	A Binary Metaheuristic Algorithm for Wrapper Feature Selection
Authors	:	Shokooh Taghian, Mohammad H. Nadimi Shahraki
Keywords	:	Classification; Feature Selection; Metaheuristic algorithm; Binary Metaheuristic Algorithm
Issue Date	:	Sep-Oct 2019
Abstract	:	The classification accuracy is strongly affected by the quality of the input features used to build a learned-model. Nowadays, datasets grow enormously both in size and number of features. One of the major difficulties confronted by huge datasets analysis is existing redundant, noisy, and irrelevant features, which may reduce the performance of the classifier. Feature selection is an important preprocessing task, which aims to select the most effective subset of features from the original dataset. Therefore, using feature selection method is essential for enhancing the classification accuracy, and reducing the complexity of the built model. In this paper, a wrapper-based binary metaheuristic algorithm for feature selection named WBMA is proposed. The proposed algorithm was compared with three well-known binary algorithms over seven classification datasets from the UCI machine learning repository. The results show that the competitive performance of the proposed algorithm in searching optimal subset and selecting the salient feature.
Page(s)	:	168-172
ISSN	:	2319-7323
Source	:	Vol. 8, No. 5