Feature Selection Based on Modified Bat Algorithm

Bin YANG; Yuliang LU; Kailong ZHU; Guozheng YANG; Jingwei LIU; Haibo YIN

doi:10.1587/transinf.2016EDP7471

Feature Selection Based on Modified Bat Algorithm

Bin YANG, Yuliang LU, Kailong ZHU, Guozheng YANG, Jingwei LIU, Haibo YIN

Full Text Views

0

Cite this

Summary :

The rapid development of information techniques has lead to more and more high-dimensional datasets, making classification more difficult. However, not all of the features are useful for classification, and some of these features may even cause low classification accuracy. Feature selection is a useful technique, which aims to reduce the dimensionality of datasets, for solving classification problems. In this paper, we propose a modified bat algorithm (BA) for feature selection, called MBAFS, using a SVM. Some mechanisms are designed for avoiding the premature convergence. On the one hand, in order to maintain the diversity of bats, they are guided by the combination of a random bat and the global best bat. On the other hand, to enhance the ability of escaping from local optimization, MBAFS employs one mutation mechanism while the algorithm trapped into local optima. Furthermore, the performance of MBAFS was tested on twelve benchmark datasets, and was compared with other BA based algorithms and some well-known BPSO based algorithms. Experimental results indicated that the proposed algorithm outperforms than other methods. Also, the comparison details showed that MBAFS is competitive in terms of computational time.

Publication: IEICE TRANSACTIONS on Information Vol.E100-D No.8 pp.1860-1869

Publication Date: 2017/08/01

Publicized: 2017/05/01

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2016EDP7471

Type of Manuscript: PAPER

Category: Pattern Recognition

Authors

Bin YANG
  Electronic Engineering Institute
Yuliang LU
  Electronic Engineering Institute
Kailong ZHU
  Electronic Engineering Institute
Guozheng YANG
  Electronic Engineering Institute
Jingwei LIU
  China Electronic System Engineering Company
Haibo YIN
  Electronic Engineering Institute

Keyword

feature selection, wrapper model, bat algorithm, premature convergence, SVM

Cite this

Copy

Bin YANG, Yuliang LU, Kailong ZHU, Guozheng YANG, Jingwei LIU, Haibo YIN, "Feature Selection Based on Modified Bat Algorithm" in IEICE TRANSACTIONS on Information, vol. E100-D, no. 8, pp. 1860-1869, August 2017, doi: 10.1587/transinf.2016EDP7471.
Abstract: The rapid development of information techniques has lead to more and more high-dimensional datasets, making classification more difficult. However, not all of the features are useful for classification, and some of these features may even cause low classification accuracy. Feature selection is a useful technique, which aims to reduce the dimensionality of datasets, for solving classification problems. In this paper, we propose a modified bat algorithm (BA) for feature selection, called MBAFS, using a SVM. Some mechanisms are designed for avoiding the premature convergence. On the one hand, in order to maintain the diversity of bats, they are guided by the combination of a random bat and the global best bat. On the other hand, to enhance the ability of escaping from local optimization, MBAFS employs one mutation mechanism while the algorithm trapped into local optima. Furthermore, the performance of MBAFS was tested on twelve benchmark datasets, and was compared with other BA based algorithms and some well-known BPSO based algorithms. Experimental results indicated that the proposed algorithm outperforms than other methods. Also, the comparison details showed that MBAFS is competitive in terms of computational time.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2016EDP7471/_p

Copy

@ARTICLE{e100-d_8_1860,
author={Bin YANG, Yuliang LU, Kailong ZHU, Guozheng YANG, Jingwei LIU, Haibo YIN, },
journal={IEICE TRANSACTIONS on Information},
title={Feature Selection Based on Modified Bat Algorithm},
year={2017},
volume={E100-D},
number={8},
pages={1860-1869},
abstract={The rapid development of information techniques has lead to more and more high-dimensional datasets, making classification more difficult. However, not all of the features are useful for classification, and some of these features may even cause low classification accuracy. Feature selection is a useful technique, which aims to reduce the dimensionality of datasets, for solving classification problems. In this paper, we propose a modified bat algorithm (BA) for feature selection, called MBAFS, using a SVM. Some mechanisms are designed for avoiding the premature convergence. On the one hand, in order to maintain the diversity of bats, they are guided by the combination of a random bat and the global best bat. On the other hand, to enhance the ability of escaping from local optimization, MBAFS employs one mutation mechanism while the algorithm trapped into local optima. Furthermore, the performance of MBAFS was tested on twelve benchmark datasets, and was compared with other BA based algorithms and some well-known BPSO based algorithms. Experimental results indicated that the proposed algorithm outperforms than other methods. Also, the comparison details showed that MBAFS is competitive in terms of computational time.},
keywords={},
doi={10.1587/transinf.2016EDP7471},
ISSN={1745-1361},
month={August},}

Copy

TY - JOUR
TI - Feature Selection Based on Modified Bat Algorithm
T2 - IEICE TRANSACTIONS on Information
SP - 1860
EP - 1869
AU - Bin YANG
AU - Yuliang LU
AU - Kailong ZHU
AU - Guozheng YANG
AU - Jingwei LIU
AU - Haibo YIN
PY - 2017
DO - 10.1587/transinf.2016EDP7471
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E100-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2017
AB - The rapid development of information techniques has lead to more and more high-dimensional datasets, making classification more difficult. However, not all of the features are useful for classification, and some of these features may even cause low classification accuracy. Feature selection is a useful technique, which aims to reduce the dimensionality of datasets, for solving classification problems. In this paper, we propose a modified bat algorithm (BA) for feature selection, called MBAFS, using a SVM. Some mechanisms are designed for avoiding the premature convergence. On the one hand, in order to maintain the diversity of bats, they are guided by the combination of a random bat and the global best bat. On the other hand, to enhance the ability of escaping from local optimization, MBAFS employs one mutation mechanism while the algorithm trapped into local optima. Furthermore, the performance of MBAFS was tested on twelve benchmark datasets, and was compared with other BA based algorithms and some well-known BPSO based algorithms. Experimental results indicated that the proposed algorithm outperforms than other methods. Also, the comparison details showed that MBAFS is competitive in terms of computational time.
ER -