Investigating the performance of MLP classifiers where limited training data are available for some classes

Parikh, CR; Pont, MJ; Li, Y; Jones, NB

Abstract

The standard implementation of the back-propagation training algorithm for multi-layer Perceptron (MLP) neural networks assumes that there are equal number of samples for training each of the required classes. Where limited training data are available for one (or more) classes, sub-optimal performance may be obtained. We have demonstrated in a previous study [Parikh et al., 1999. Proceedings of Condition Monitoring 1999, Swansea, UK] that, where unequal training class cannot be avoided, performance of the classifier may be substantially improved by duplicating the available patterns in the smaller class. In this study, we investigate whether the addition of random noise to the `duplicated' training patterns will further improve the classification performance. In the study conducted here, we conclude that the addition of noise does not give a consistent improvement in performance.

Citation

Parikh, C., Pont, M., Li, Y., & Jones, N. (1999, July). Investigating the performance of MLP classifiers where limited training data are available for some classes. Presented at Recent Advances in Soft Computing Techniques and Applications, Leicester, UK

• Presentation Conference Type: Other
• Conference Name: Recent Advances in Soft Computing Techniques and Applications
• Conference Location: Leicester, UK
• Start Date: Jul 1, 1999
• End Date: Jul 2, 1999
• Publication Date: Jul 1, 2000
• Deposit Date: Jul 27, 2015
• Series Title: Advances in Soft Computing
• Additional Information: Event Type : Conference