A novel support vector machine ensemble based on subtractive clustering analysis

Authors:
Cuiru Wang;Hejin Yuan;Jun Liu;Tao Zhou;Huiling Lu
Affiliations:
Department of Computer Science, North China Electric Power University, Baoding, Hebei, China;Department of Computer Science, North China Electric Power University, Baoding, Hebei, China;Department of Computer Science, North China Electric Power University, Baoding, Hebei, China;Department of Maths, Shaanxi University of Technology, Hanzhong, Shaanxi, China;Department of Computer, Shaanxi University of Technology, Hanzhong , Shaanxi, China
Venue:
PAKDD'07 Proceedings of the 11th Pacific-Asia conference on Advances in knowledge discovery and data mining
Year:
2007

Citing 3
Cited 1

Bagging predictors

Machine Learning
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
Automatic news audio classification based on selective ensemble SVMs

ISNN'05 Proceedings of the Second international conference on Advances in neural networks - Volume Part II

Hierarchical annotation of medical images

Pattern Recognition

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper put forwards a novel support vector machine ensemble construction method based on subtractive clustering analysis. Firstly, the training samples are clustered into several clusters according to their distribution with subtractive clustering algorithm. Then small quantities of representative instances from them are chosen as training subsets to construct support vector machine components. At last, the base classifiers' outputs are aggregated to obtain the final decision. Experiment results on UCI datasets show that the SVM ensemble generated by our method has higher classification accuracy than Bagging, Adaboost and k-fold cross validation algorithms.