Scaling the kernel function based on the separating boundary in input space: A data-dependent way for improving the performance of kernel methods
作者： Jiancheng Sun , Xiaohe Li , Yong Yang , Jianguo Luo , Yaohui Bai
DOI: 10.1016/J.INS.2011.08.028
关键词: Tree kernel 、 Radial basis function kernel 、 String kernel 、 Kernel embedding of distributions 、 Artificial intelligence 、 Polynomial kernel 、 Kernel (statistics) 、 Kernel principal component analysis 、 Variable kernel density estimation 、 Pattern recognition 、 Mathematics
摘要: The performance of a kernel method often depends mainly on the appropriate choice function. In this study, we present data-dependent for scaling function so as to optimize classification methods. Instead finding support vectors in feature space, first find region around separating boundary input and subsequently scale correspondingly. It is worth noting that proposed does not require training step enable specified algorithm can be applied various Experimental results using both artificial real-world data are provided demonstrate robustness validity method.
elsevier.com
sci-hub.se 参考文章(36)
TH Cormen, RL Rivest, CE Leiserson, C Stein, Introduction to Algorithms, 2nd edition. ,(2001)
Si Wu, Shun-Ichi Amari, Conformal Transformation of Kernel Functions: A Data-Dependent Way to Improve Support Vector Machine Classifiers Neural Processing Letters. ,vol. 15, pp. 59- 67 ,(2002) , 10.1023/A:1013848912046
Christopher J. C. Burges, Geometry and invariance in kernel based methods Advances in kernel methods. pp. 89- 116 ,(1999)
Hyunjung Shin, Sungzoon Cho, Neighborhood Property--Based Pattern Selection for Support Vector Machines Neural Computation. ,vol. 19, pp. 816- 855 ,(2007) , 10.1162/NECO.2007.19.3.816
Daniel S. Yeung, Defeng Wang, Wing W. Y. Ng, Eric C. C. Tsang, Xizhao Wang, Structured large margin machines: sensitive to data distributions Machine Learning. ,vol. 68, pp. 171- 200 ,(2007) , 10.1007/S10994-007-5015-9
Joshua B Tenenbaum, Vin de Silva, John C Langford, A Global Geometric Framework for Nonlinear Dimensionality Reduction Science. ,vol. 290, pp. 2319- 2323 ,(2000) , 10.1126/SCIENCE.290.5500.2319
Zhihua Zhang, Dit-Yan Yeung, James T. Kwok, Bayesian inference for transductive learning of kernel matrix using the Tanner-Wong data augmentation algorithm Twenty-first international conference on Machine learning - ICML '04. pp. 118- ,(2004) , 10.1145/1015330.1015368
Cheng-Hsuan Li, Chin-Teng Lin, Bor-Chen Kuo, Hui-Shan Chu, None, An automatic method for selecting the parameter of the RBF kernel function to support vector machines 2010 IEEE International Geoscience and Remote Sensing Symposium. pp. 836- 839 ,(2010) , 10.1109/IGARSS.2010.5649251
Francis R Bach, Gert RG Lanckriet, Michael I Jordan, None, Multiple kernel learning, conic duality, and the SMO algorithm Twenty-first international conference on Machine learning - ICML '04. pp. 6- ,(2004) , 10.1145/1015330.1015424
Qiang He, Congxin Wu, Separating theorem of samples in Banach space for support vector machine learning International Journal of Machine Learning and Cybernetics. ,vol. 2, pp. 49- 54 ,(2011) , 10.1007/S13042-011-0013-4