Classification Using Radial Basis Function Networks with Uncertain Weights

Graeme Manson; Gareth Pierce; Keith Worden; Daley Chetwynd

doi:10.4028/www.scientific.net/KEM.293-294.135

Paper Titles

Comparison of Two Software Tools for Damage Identification: Gradient-Based vs. Case-Based Approach
p.103

Identification of Impact Force for Base on Higher Order Wigner Distribution
p.111

Damage Identification of a Mechanical Structure with a Non-Minimum Phase Based on Singular Value Decomposition
p.119

Wavelet Transform Domain Filter and Its Application in Incipient Fault Prognosis
p.127

Classification Using Radial Basis Function Networks with Uncertain Weights
p.135

Comparison of Different Methods of Ritz Vectors Extraction
p.143

Considerations for Practical Neural Network Application to a Damage Detection Problem
p.151

Impact Load Identification Based on Local Measurements
p.159

Extracting Gear Fault Features Using Maximal Bispectrum
p.167

HomeKey Engineering MaterialsKey Engineering Materials Vols. 293-294Classification Using Radial Basis Function...

Classification Using Radial Basis Function Networks with Uncertain Weights

Abstract:

This paper considers the performance of radial basis function neural networks for the purpose of data classification. The methods are illustrated using a simple two class problem. Two techniques for reducing the rate of misclassifications, via the introduction of an “unable to classify” label, are presented. The first of these considers the imposition of a threshold value on the classifier outputs whilst the second considers the replacement of the crisp network weights with interval ranges. Two network training techniques are investigated and it is found that, although thresholding and uncertain weights give similar results, the level of variability of network performance is dependent upon the training approach

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Key Engineering Materials (Volumes 293-294)

Pages:

135-142

DOI:

https://doi.org/10.4028/www.scientific.net/KEM.293-294.135

Citation:

Cite this paper

Online since:

September 2005

Authors:

Graeme Manson, Gareth Pierce, Keith Worden, Daley Chetwynd

Keywords:

Damage Detection, Interval Arithmetic, Neural Network (NN), Radial Basis Function (RBF), Uncertainty

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] H. Ishibuchi and H. Tanaka: Fuzzy Sets and Systems, 50 (1992), pp.257-265.

Google Scholar

[2] G. Manson, S.G. Pierce, K. Worden, and D. Chetwynd: Classification using linear models with uncertain weights, EURODYN Conference, Paris, France, (2005).

Google Scholar

[3] G. Manson, K. Worden, and D. Allman: Journal of Sound and Vibration, 259(2) (2003), pp.345-363.

Google Scholar

[4] G. Manson, K. Worden, and D. Allman.: Journal of Sound and Vibration, 259(2) (2003), pp.365-385.

Google Scholar

[5] G. Manson: Identifying damage sensitive, environment insensitive features for damage detection, Proceedings of International Conference on Identification in Engineering Systems, Swansea (2002), pp.187-197.

Google Scholar

[6] I.T. Nabney: NETLAB - Algorithms for pattern recognition (Springer-Verlag, 2002).

Google Scholar

[7] R. Moore: Interval Analysis, (Prentice-Hall, Englewood Cliffs, NJ. 1966).

Google Scholar