University of Stirling    STORRE: Stirling Online Research Repository University Circle Images   Research Led, Student Focused  
 

STORRE >
School of Natural Sciences >
Computing Science and Mathematics >
Computing Science and Mathematics Journal Articles >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1893/1654

Appears in Collections:Computing Science and Mathematics Journal Articles
Peer Review Status: Refereed
Title: Feature subset selection in large dimensionality domains
Author(s): Gheyas, Iffat A
Smith, Leslie
Contact Email: l.s.smith@cs.stir.ac.uk
Keywords: feature subset selection
high dimensional datasets
Issue Date: Jan-2010
Publisher: Elsevier
Citation: Gheyas IA & Smith L (2010) Feature subset selection in large dimensionality domains, Pattern Recognition, 43 (1), pp. 5-13.
Abstract: Searching for an optimal feature subset from a high dimensional feature space is known to be an NP-complete problem. We present a hybrid algorithm, SAGA, for this task. SAGA combines the ability to avoid being trapped in a local minimum of Simulated Annealing with the very high rate of convergence of the crossover operator of Genetic Algorithms, the strong local search ability of greedy algorithms and the high computational efficiency of Generalized Regression Neural Networks. We compare the performance over time of SAGA and well-known algorithms on synthetic and real datasets. The results show that SAGA outperforms existing algorithms.
Type: Journal Article
URI: http://hdl.handle.net/1893/1654
URL: http://www.sciencedirect.com/science/journal/00313203
DOI Link: http://dx.doi.org/10.1016/j.patcog.2009.06.009
Rights: Published in Pattern Recognition by Elsevier.
Affiliation: University of Stirling
Computing Science - CSM Dept

Files in This Item:

File Description SizeFormat
PR_3589_corrected.pdf478.04 kBAdobe PDFView/Open


This item is protected by original copyright

Recommend this item

Items in the Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

If you believe that any material held in STORRE infringes copyright, please contact library@stir.ac.uk providing details and we will remove the Work from public display in STORRE and investigate your claim.

 

Valid XHTML 1.0! Repository Service Operated by Information Services, University of Stirling
Powered by DSpace Software Copyright © 2002-2010  Duraspace - Feedback