Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/16197
Title: Quantitative study on effect of center selection to RBFNN classification performance
Authors: Ng, WY
Yeung, DS
Cloete, I
Keywords: Pattern classification
Radial basis function networks
Stochastic processes
Issue Date: 2004
Publisher: IEEE
Source: 2004 IEEE International Conference on Systems, Man and Cybernetics, 10-13 October 2004, v. 4, p. 3692-3697 How to cite?
Abstract: In pattern classification problems using a RBFNN classifier, the selection of the number of clusters and their corresponding centers influences the network's ability to generalize unseen data. In this paper, we evaluate different RBFNN architectures by a quantitative measure - RBFNN sensitivity measure, which is defined as the absolute expectation plus standard deviation of network output perturbations with respect to input perturbations. Numerical comparisons of a number of different RBFNN architectures are given using two of UCI datasets. The experiments show that the sensitivity measure would be correlated to the testing error for the unseen samples and simpler classification problem may have smaller sensitivity measure.
URI: http://hdl.handle.net/10397/16197
ISBN: 0-7803-8566-7
ISSN: 1062-922X
DOI: 10.1109/ICSMC.2004.1400917
Appears in Collections:Conference Paper

Access
View full-text via PolyU eLinks SFX Query
Show full item record

SCOPUSTM   
Citations

4
Last Week
0
Last month
Citations as of Jul 10, 2018

WEB OF SCIENCETM
Citations

3
Last Week
0
Last month
0
Citations as of Jul 12, 2018

Page view(s)

52
Last Week
0
Last month
Citations as of Jul 23, 2018

Google ScholarTM

Check

Altmetric


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