A new evolutionary neural network classifier

Authors:
Arit Thammano;Asavin Meengen
Affiliations:
Faculty of Information Technology, King Mongkut's Institute of Technology Ladkrabang, Bangkok, Thailand;Faculty of Information Technology, King Mongkut's Institute of Technology Ladkrabang, Bangkok, Thailand
Venue:
PAKDD'05 Proceedings of the 9th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining
Year:
2005

Citing 3
Cited 0

Fuzzy RuleNet: an artificial neural network model for fuzzy classification

SAC '94 Proceedings of the 1994 ACM symposium on Applied computing
Data mining: concepts and techniques

Data mining: concepts and techniques
A Compact and Accurate Model for Classification

IEEE Transactions on Knowledge and Data Engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes two new concepts: (1) the new evolutionary algorithm and (2) the new approach to deal with the classification problems by applying the concepts of the fuzzy c-means algorithm and the evolutionary algorithm to the artificial neural network. During training, the fuzzy c-means algorithm is initially used to form the clusters in the cluster layer; then the evolutionary algorithm is employed to optimize those clusters and their parameters. During testing, the class whose cluster node returns the maximum output value is the result of the prediction. This proposed model has been benchmarked against the standard backpropagation neural network, the fuzzy ARTMAP, C4.5, and CART. The results on six benchmark problems are very encouraging.