Title :
Selectively ensembling neural classifiers
Author :
Zhou, Zhi-Hua ; Wu, Jianxin ; Tang, Wei ; Chen, Zhao-Qian
Author_Institution :
Nat. Lab. for Novel Software Technol., Nanjing Univ., China
fDate :
6/24/1905 12:00:00 AM
Abstract :
Ensembling neural classifiers can significantly improve the generalization ability of classification systems. In this paper, GASEN, a genetic algorithm based selective ensemble method, that has been shown to be excellent in ensembling neural regressors, is applied to neural classifiers. Experiments on four large data sets show that this method can generate ensembles of neural classifiers with stronger generalization ability than those generated by Bagging, Adaboost, or Arc-x4
Keywords :
classification; generalisation (artificial intelligence); genetic algorithms; neural nets; GASEN; classification; generalization; genetic algorithm; heuristics; neural classifiers; neural network ensemble; neural regressors; selective ensemble method; Bagging; Biomedical optical imaging; Character recognition; Face recognition; Genetic algorithms; Handwriting recognition; Image recognition; Laboratories; Neural networks; Optical character recognition software;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007723
