Browse > Article
http://dx.doi.org/10.3745/KIPSTB.2005.12B.1.025

Performance comparison of SVM and neural networks for large-set classification problems  

Lee Jin-Seon (우석대학교 컴퓨터공학과)
Kim Young-Won (전북대학교 대학원 컴퓨터정보학과)
Oh Il-Seok (전북대학교 전자정보공학부)
Publication Information
The KIPS Transactions:PartB / v.12B, no.1, 2005 , pp. 25-30 More about this Journal
Abstract
In this paper, we analyzed and compared the performances of modular FFMLP(feedforward multilayer perceptron) and SVUT(Support Vector Machine) for the large-set classification problems. Overall, SVM dominated modular FFMLP in the correct recognition rate and other aspects Additionally, the recognition rate of SVM degraded more slowly than neural network as the number of classes increases. The trend of the recognition rates depending on the rejection rate has been analyzed. The parameter set of SVM(kernel functions and related variables) has been identified for the large-set classification problems.
Keywords
Large-set Classification; SVM; Modular Feedforward Neural Network; Performance Comparison;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Hee-Heon Song and Seong-Whan Lee, 'A self-organizing neural tree for large-set pattern classification,' IEEE Transactions on Neural Networks, Vol.9, No.3, pp.369- 380, 1998   DOI   ScienceOn
2 Il-Seok Oh and Ching Y. Suen, 'A class-modular feedforward neural network for handwriting recognition,' Pattern Recognition, Vol.35, No.1, pp.229-244, 2002   DOI   ScienceOn
3 C. J. C. Burges, A tutorial on support vector machines for pattern recognition, Data Mining and Knowledge Discovery, Vol.2, pp.121-167, 1998   DOI   ScienceOn
4 http://svmlight.joachims.org
5 R. O. Duda, P. E. Hart, and D. G. Stork, Pattern Classification, 2nd Ed., Wiley- Interscience, 2001
6 Hahn-Ming Lee, Chin-Chou Lin, and Jyh-Ming Chen, 'A preclassification method for handwritten Chinese character recognition via fuzzy rules and SEART neural net,' International Journal of Pattern Recognition and Artificial Intelligence, Vo1.12, No.6, pp.743-761, 1998   DOI   ScienceOn