Research on Parameter Selection of Support Vector Regression

Chen Wu; Hong Yao; Jun Du; Jiu Long Jiang

doi:10.4028/www.scientific.net/AMM.344.219

Paper Titles

Chaos Identification and Suppression in an Indirect Field-Oriented Control of Induction Motor Drive Based on 0-1 Method
p.194

Vector Control of Brushless Doubly-Fed Machine for High Power Stand-Alone Power Generation
p.199

Research on Two Kinds of Parameter Identification Methods for Short-Term Time-Varying System
p.205

Supervising Cognitive Tele-Rehabilitation with a Virtual Therapist
p.213

Research on Parameter Selection of Support Vector Regression
p.219

Research of Blurred Face Image Detection Based on DCT and Edge Detection Algorithm
p.226

Fast Stereo Video Coding Based on C-Means Fuzzy Hopfield Neural Network and AVS
p.230

An Improvement Measure to AC-BM Pattern Matching Algorithm in Intrusion Defense System
p.234

A Fast Interpolation Algorithm Based on the Fractal in the Digital Speckle Sub-Pixel Displacement Measurement
p.238

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vol. 344Research on Parameter Selection of Support Vector...

Research on Parameter Selection of Support Vector Regression

Abstract:

Parameters of support vector regression (SVR) have a great impact on complexity, training accuracy and prediction accuracy of the model. To solve the problem that excessive pursuit of training accuracy will decrease the generalization ability of model, a method of parameter optimization based on artificial fish swarm algorithm (AFSA) is put forward. A part of predictive sample is used to calculate error, and build fitness function according to it, which set up the feedback from prediction to training and avoid overfitting. The simulation results show that the method decreases the prediction error of SVR, and improve the generalization ability.

You might also be interested in these eBooks

Advanced Research on Applied Mechanics, Mechatronics and Intelligent System

View Preview

Info:

Periodical:

Applied Mechanics and Materials (Volume 344)

Pages:

219-225

DOI:

https://doi.org/10.4028/www.scientific.net/AMM.344.219

Citation:

Cite this paper

Online since:

July 2013

Authors:

Chen Wu, Hong Yao, Jun Du, Jiu Long Jiang

Keywords:

Artificial Fish-Swarm Algorithm (AFSA), Overfitting, Parameter Optimization, Support Vector Regression (SVR)

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Vapnik V N, The Nature of Statistical Learning Theory, New York: Springer Verlag, (1999).

Google Scholar

[2] Smola A J, Scholkope B, A tutorial on support vector regression , Statistics and Computing, 2004, 14: 199-222.

DOI: 10.1023/b:stco.0000035301.49549.88

Google Scholar

[3] Gan X S, Zhang H C, Cheng Y M, et al, Aerodynamic parameter fitting based on robust least squares support vector machines, Computer Engineering and Applications, 2007, 43(31): 233-235.

DOI: 10.1109/snpd.2007.173

Google Scholar

[4] T. Hofmann, B. Scholkopf, and A.J. Smola. A Review of Kernel Methods in Machine Learning, Technical Report No. 156, (2006).

Google Scholar

[5] YANG X X, ZHANG W H. A new strategy for improving particle swarm optimization, International Conference on Intelligent Computing Technology and Automation. Changsha, 2009: 228- 232.

Google Scholar

[6] LI X L, SHAO Z J, QIAN J X. An optimizing method based on autonomous animate: fish-swarm algorithm, System Engineering Theory and Practice, 2002, 22, (11): 32-38.

Google Scholar

[7] LI X L, SHAO Z J, QIAN J X. Studies on Artificial Fish-Swarm Optimization Algorithm based on Decomposition and Coordination Techniques. Journal of Circuits and Systems, 2003, 23(8): 1-6.

Google Scholar