New Feature Selection Method Based on SVM-RFE

Xi Wang; Qiang Li; Zhi Hong Xie

doi:10.4028/www.scientific.net/AMR.926-930.3100

Paper Titles

Modeling and Analysis of the Ground Experiment for Restricted Three-Body Problem
p.3084

Modeling and Simulating for Queuing System with QuickPass
p.3088

Multi-Attribute Decision Making Based on the Similarity Degree of Ternary Interval Numbers
p.3092

Multi-Field Coupling Simulation on Quench Process in Superconducting Magnet
p.3096

New Feature Selection Method Based on SVM-RFE
p.3100

Numerical Algorithm of the Fractional Diffusion Equation Based on Sinc-Chebyshev Collocation
p.3105

Numerical Analysis of Wave Overtopping Failure of Booms with SPH
p.3109

Numerical Methods for Solving Implied Volatility in Option Pricing
p.3115

Numerical Simulation of Loading Test on Rubble Bed by Granular Discrete Element Method
p.3119

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 926-930New Feature Selection Method Based on SVM-RFE

New Feature Selection Method Based on SVM-RFE

Abstract:

This article analyzed the defects of SVM-RFE feature selection algorithm, put forward new feature selection method combined SVM-RFE and PCA. Firstly, get the best feature subset through the method of cross validation of k based on SVM-RFE. Then, the PCA decreased the dimension of the feature subset and got the independent feature subset. The independent feature subset was the training and testing subset of SVM. Make experiments on five subsets of UCI, the results indicated that the training and testing time was shortened and the recognition accuracy rate of the SVM was higher.

You might also be interested in these eBooks

Progress in Applied Sciences, Engineering and Technology

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 926-930)

Pages:

3100-3104

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.926-930.3100

Citation:

Cite this paper

Online since:

May 2014

Authors:

Xi Wang*, Qiang Li, Zhi Hong Xie

Keywords:

Feature Selection, Principal Component Analysis (PCA), Recursive Feature Elimination, Support Vector Machine (SVM)

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

* - Corresponding Author

References

[1] Yudong Zhang, Yuankai Huo, Leman Wu, Dimensionality reduction techniques and methods[J]. Journal of ordnance in Sichuan, 2010, 31(10).

Google Scholar

[2] Yujiao Jiang, Xizodan Wang, a feature selection approach base on PCA and ReliefF[J]. Computer engineering and application, 2010, 46(26).

Google Scholar

[3] Guyon I, Weston J. Gene Selection for Cancer Classification using Support Vector Machines [J]. Machine Learning, 2002, 46(1-3): 389-442.

Google Scholar

[4] Zhou Xin, Tuck DP. MSVM-RFE: extensions of SVM-RFE for multiclass gene selection on DNA microarray data [J]. Bi oinformatics, 2007, 23 (9) : 1106 - 1114.

DOI: 10.1093/bioinformatics/btm036

Google Scholar

[5] V. Vapnik. The Nature of Statistics Learning Theory [M]. New York: Springer Verlag, (1995).

Google Scholar

[6] Xuegong Zhang. Statistical learning theory and support vector machine[J], Journal of automatization, 2000, 26(1), 32-42.

Google Scholar

[7] Zhaoqi Bian, Xuegong Zhang, Pattern recognition[M]. Publishing house of Tsinghua University2001.

Google Scholar

[8] Blake C L, Merz C J. UCI Repository of Machine Learning Databases [EB/OL].

Google Scholar