Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13] [ 14] [ 15]


Journal of Information Science and Engineering, Vol. 27 No. 5, pp. 1667-1686 (September 2011)

Feature Subset Selection Based on Bio-Inspired Algorithms*

Department of Computer Science and Engineering
Sogang University
Seoul, 121-742 Korea

Many feature subset selection algorithms have been proposed and discussed for years. However, the problem of finding the optimal feature subset from full data still remains to be a difficult problem. In this paper, we propose novel methods to find the relevant feature subset by using biologically-inspired algorithms such as Genetic Algorithm and Particle Swarm Optimization. We also propose a variant of the approach considering the significance of each feature. We verified the performance of the proposed methods by experiments with various real-world datasets. Our feature selection methods based on the biologically-inspired algorithms produced better performance than other methods in terms of the classification accuracy and the feature relevance. In particular, the modified method considering feature significance demonstrated even more improved performance.

Keywords: genetic algorithm, particle swarm optimization, feature redundancy and relevance, wrapper approach, inductive learning algorithm

Full Text () Retrieve PDF document (201109_10.pdf)

Received December 7, 2009 revised March 29 & November 1, 2010; accepted December 15, 2010.
Communicated by Chih-Jen Lin.
* This paper was supported by the Special Research Grant of Sogang University to Jihoon Yang.
+ Corresponding author.