An Effective Method for Combining Kernels with Class Separability

Ting Hua Wang; Qiong Zhang; Hai Hui Xie; Jun Ting Chen

doi:10.4028/www.scientific.net/AMM.220-223.2224

Paper Titles

Object Annotation Based on Middle Semantic Manifold
p.2204

Motion Compensation Based on Keystone Transform and Wigner-Hough Transform
p.2208

Object Remarkable Features Extraction Strategy Based on Imagery Conceptual Network
p.2213

BOU-Based Cycle Determination for Different Kinds of Mechanical Vibration Signals
p.2217

An Effective Method for Combining Kernels with Class Separability
p.2224

Feature Extraction of Nonstationarity Vibration Signal Based on Wavelet Decomposition
p.2228

Harmonic and Inter-Harmonic Detection Based on FFT and Nonlinear Optimization
p.2235

Research of Electronic Nose Pattern Recognition Algorithm Based on SVM
p.2244

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 220-223An Effective Method for Combining Kernels with...

An Effective Method for Combining Kernels with Class Separability

Abstract:

We propose a simple but effective method to determine the kernel weights for convex combination of multiple kernels. The key property of the proposed method is that it adopts a class separability criterion as the evaluation function to measure the goodness of the individual kernel. Based on the principle of class separability, we assign a weight to each kernel that is proportional to the quality of the kernel. Experimental results on Image Segmentation data set show the proposed method can improve accuracy in comparison with that using a single kernel or uniformly-combined kernel.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volumes 220-223)

Pages:

2224-2227

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.220-223.2224

Citation:

Cite this paper

Online since:

November 2012

Authors:

Ting Hua Wang, Qiong Zhang, Hai Hui Xie, Jun Ting Chen

Keywords:

Class Separability, Kernel Method, Multiclass Classification, Multiple Kernel Learning (MKL), Support Vector Machine (SVM)

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] J. Shawe-Taylor and N. Cristianini. Kernel methods for pattern analysis. Cambridge University Press, Cambridge (2004).

Google Scholar

[2] V. Vapnik. The nature of Statistical Learning Theory. Springer-Verlag, New York (1995).

Google Scholar

[3] G.R.G. Lanckriet, N. Cristianini, P. Bartlett, L.E. Ghaoui, and M.I. Jordan. Learning the kernel matrix with semidefinite programming. Journal of Machine Learning Research 5 (2004): 27-72.

Google Scholar

[4] F.R. Bach, G.R.G. Lanckriet, and M.I. Jordan. Multiple kernel learning, conic duality, and the SMO algorithm. In: Proceedings of the 21st International Conference on Machine Learning (2004): 41-48.

DOI: 10.1145/1015330.1015424

Google Scholar

[5] M. Gönen and E. Alpayın. Multiple kernel learning algorithms. Journal of Machine Learning Research 12 (2011): 2211-2268.

Google Scholar

[6] A. Zien and C.S. Ong. Multiclass multiple kernel learning. In: Proceedings of the 24th International Conference on Machine Learning (2007): 1191–1198.

DOI: 10.1145/1273496.1273646

Google Scholar

[7] L. Wang. Feature selection with kernel class separability. IEEE Transactions on Pattern Analysis and Machine Intelligence 30 (9) (2008): 1534-1546.

DOI: 10.1109/tpami.2007.70799

Google Scholar

[8] A. Frank and A. Asuncion. UCI Machine Learning Repository. http://archive.ics.uci.edu/ml/, (2010).

Google Scholar