Embedded feature ranking for ensemble MLP classifiers

Terry Windeatt; Rakkrit Duangsoithong; Raymond Smith

doi:10.1109/TNN.2011.2138158

Embedded feature ranking for ensemble MLP classifiers

IEEE Trans Neural Netw. 2011 Jun;22(6):988-94. doi: 10.1109/TNN.2011.2138158. Epub 2011 May 19.

Authors

Terry Windeatt¹, Rakkrit Duangsoithong, Raymond Smith

Affiliation

¹ Centre for Vision Speech and Signal Processing, Faculty of Electronics and Physical Sciences, University of Surrey, Guildford Surrey, UK. t.windeatt@surrey.ac.uk

PMID: 21606023
DOI: 10.1109/TNN.2011.2138158

Abstract

A feature ranking scheme for multilayer perceptron (MLP) ensembles is proposed, along with a stopping criterion based upon the out-of-bootstrap estimate. To solve multi-class problems feature ranking is combined with modified error-correcting output coding. Experimental results on benchmark data demonstrate the versatility of the MLP base classifier in removing irrelevant features.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Algorithms*
Computer Simulation
Models, Theoretical*
Neural Networks, Computer*
Pattern Recognition, Automated / methods*