Earticle

초록

영어

Support vector machine (SVM) is a kind of machine learning method, but the selection of parameters has important effects on the generalization ability of SVMs. In this study, the relation between the error penalty parameter C, kernel parameter σ and the generalization ability of SVMs is discussed. Parameter C adjusts the similarity among within-class members, while parameter σ adjusts the similarity between classes. Moreover, C and σ balances each other mutually within a certain range, which forms a fan-shaped optional parameter distribution region. The optimal parameter area should be located near the center of the sector where both C and σ are small. According to this, a method is suggested to first search a suitable area with coarse grids, and then determine the optimal parameter within the area with a fine bilinear grid. Experimental results show that the new parameter selection method can not only avoid local optima, and thus excluding the cases in which C and σ are big and unstable, but also can be extremely fast in searching process. Compared with other parameter selection methods, the performance of SVMs cannot be influenced, or even better in some cases.

목차

Abstract
 1. Introduction
 2. The Relation between the Performance of SVMs with Gaussian Kernels and Parameter C and σ
  2.1. The Relation between Error Penalty Parameter C and Generalization Ability of SVMs
  2.2. The Relation between the Kernel Parameter σ and the Generalization Ability of SVMs
  2.3. Rethinking of the Relation between (C, σ) and the Generalization Ability of SVMs
  2.4. The Optimal Ranges of (C, σ)
 3. The Improvements of Parameter Selection Method for SVMs with Gaussian Kernel
 4. The Relation between the Distribution Characters of 2(,)Cσ and Sample Dimensionality
 5．Conclusion
 Acknowledgements
 References

키워드

저자

• Zhigang Yan [ School of Environmental Science and Spatial Informatics, China University of Mining and Technology, Jiangsu Key Laboratory of Resource and Environmental Information Engineering, China University of Mining and Technology, Xuzhou City, Jiangsu Province, 221116, P.R.China ] Corresponding author
• Yuanxuan Yang [ School of Environmental Science and Spatial Informatics, China University of Mining and Technology, Jiangsu Key Laboratory of Resource and Environmental Information Engineering, China University of Mining and Technology, Xuzhou City, Jiangsu Province, 221116, P.R.China ]
• Yunjing Ding [ School of Environmental Science and Spatial Informatics, China University of Mining and Technology, Jiangsu Key Laboratory of Resource and Environmental Information Engineering, China University of Mining and Technology, Xuzhou City, Jiangsu Province, 221116, P.R.China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Signal Processing, Image Processing and Pattern Recognition

• 간기

  격월간

• pISSN

  2005-4254

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605