A Multiple Evolutionary Neural Network Classifier Based on Niche Genetic Algorithm

Author

Wu, Jiang ; Tang, Changjie ; Li, Taiyong ; Jiang, Yue ; Ye, Shangyu ; Lin, Xun ; Zuo, Jie

Author_Institution

Sch. of Comput. Sci., Sichuan Univ., Chengdu

Volume

3

fYear

2008

fDate

18-20 Oct. 2008

Firstpage

405

Lastpage

409

Abstract

Classification is important in data mining. In this paper, a multiple evolutionary neural network classifier based on niche genetic algorithm (MNC-NG) is presented, which establishes classifiers by a group of three-layer feed-forward neural networks with high accuracy and good diversity. The neural networks are trained by niche genetic algorithm based on clustering. The class label of the identifying data can first be evaluated by each neural network, and the final classification result is obtained according to the dynamic voting rule. Experimental results on 6 data sets show that MNC-NG increases the predictive accuracy by 5.6%, 5.5% and 8.5% respectively compared with BP, GA and LM training methods and by 6.0%, 6.1% and 4.0% compared with Naive Bayesian classifier, C4.5 and SVM.

Keywords

data mining; evolutionary computation; feedforward neural nets; genetic algorithms; pattern classification; data mining; multiple evolutionary neural network classifier; niche genetic algorithm; three-layer feedforward neural networks; Accuracy; Bayesian methods; Data mining; Feedforward neural networks; Feedforward systems; Genetic algorithms; Neural networks; Support vector machine classification; Support vector machines; Voting;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation, 2008. ICNC '08. Fourth International Conference on

Conference_Location

Jinan

Print_ISBN

978-0-7695-3304-9

Type

conf

DOI

10.1109/ICNC.2008.602

Filename

4667170