DocumentCode :
1582941
Title :
A Study of a Multi-class Classification Algorithm of SVM Combined with ART
Author :
Wang, Anna ; Yuan, Wenjing ; Liu, Junfang ; Wang, Qinwan ; Yu, Zhiguo
Author_Institution :
Northeastern Univ., Shenyang
Volume :
1
fYear :
2007
Firstpage :
59
Lastpage :
63
Abstract :
This paper provides a novel multi-class classification algorithm, which combines adaptive resonance theory with support vector machine principle. It improves the one-against-one classification of support vector machine. The algorithm adopts adaptive resonance theory network to fuse the classifiers´ results and does not adopt voting principle. When the outputs of classifiers approach zero and the algorithm gets the same votes, it avoids the fusing errors coming from voting principle. We use this algorithm in fault diagnosis of power line network and give accurate results of classification.
Keywords :
adaptive resonance theory; fault diagnosis; pattern classification; power cables; power system analysis computing; power system faults; support vector machines; adaptive resonance theory; fault diagnosis; multiclass classification algorithm; power line network; support vector machine; Classification algorithms; Educational institutions; Hydrogen; Quadratic programming; Resonance; Risk management; Subspace constraints; Support vector machine classification; Support vector machines; Voting;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Natural Computation, 2007. ICNC 2007. Third International Conference on
Conference_Location :
Haikou
Print_ISBN :
978-0-7695-2875-5
Type :
conf
DOI :
10.1109/ICNC.2007.147
Filename :
4344154
Link To Document :
بازگشت