Selecting salient features for classification committees

Authors:
Antanas Verikas;Marija Bacauskiene;Kerstin Malmqvist
Affiliations:
Intelligent Systems Laboratory, Halmstad University, Halmstad, Sweden and Department of Applied Electronics, Kaunas University of Technology, Kaunas, Lithuania;Department of Applied Electronics, Kaunas University of Technology, Kaunas, Lithuania;Intelligent Systems Laboratory, Halmstad University, Halmstad, Sweden
Venue:
ICANN/ICONIP'03 Proceedings of the 2003 joint international conference on Artificial neural networks and neural information processing
Year:
2003

Citing 7
Cited 1

Optimal brain damage

Advances in neural information processing systems 2
Merging back-propagation and Hebbian learning rules for robust classifications

Neural Networks
A decision-theoretic generalization of on-line learning and an application to boosting

Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
Boosting regression estimators

Neural Computation
Soft combination of neural classifiers: a comparative study

Pattern Recognition Letters
Pruning Adaptive Boosting

ICML '97 Proceedings of the Fourteenth International Conference on Machine Learning
Neural-network feature selector

IEEE Transactions on Neural Networks

Selecting salient features for classification based on neural network committees

Pattern Recognition Letters

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a neural network based approach for identifying salient features for classification in neural network committees. Our approach involves neural network training with an augmented cross-entropy error function. The augmented error function forces the neural network to keep low derivatives of the transfer functions of neurons of the network when learning a classification task. Feature selection is based on two criteria, namely the reaction of the cross-validation data set classification error due to the removal of the individual features and the diversity of neural networks comprising the committee. The algorithm developed removed a large number of features from the original data sets without reducing the classification accuracy of the committees. By contrast, the accuracy of the committees utilizing the reduced feature sets was higher than those exploiting all the original features.