Simultaneous feature selection and parameters optimization for SVM by immune clonal algorithm

Authors:
Xiangrong Zhang;Licheng Jiao
Affiliations:
National Key Lab for Radar Signal Processing, Institute of Intelligent Information Processing, Xidian University, Xi'an, China;National Key Lab for Radar Signal Processing, Institute of Intelligent Information Processing, Xidian University, Xi'an, China
Venue:
ICNC'05 Proceedings of the First international conference on Advances in Natural Computation - Volume Part II
Year:
2005

Citing 9
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
Choosing Multiple Parameters for Support Vector Machines

Machine Learning
Feature Selection for Support Vector Machines by Means of Genetic Algorithms

ICTAI '03 Proceedings of the 15th IEEE International Conference on Tools with Artificial Intelligence
Hybrid Genetic Algorithms for Feature Selection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Support Vector Machines: Theory and Applications (Studies in Fuzziness and Soft Computing)

Support Vector Machines: Theory and Applications (Studies in Fuzziness and Soft Computing)
Bounds on Error Expectation for Support Vector Machines

Neural Computation
Selective SVMs ensemble driven by immune clonal algorithm

EC'05 Proceedings of the 3rd European conference on Applications of Evolutionary Computing
Dimensionality reduction using genetic algorithms

IEEE Transactions on Evolutionary Computation
Data dimensionality reduction with application to simplifying RBF network structure and improving classification performance

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics

Quantified Score

Hi-index	0.00

Visualization

Abstract

The problems of feature selection and automatically tuning parameters for SVM are considered at the same time. It is reasonable because the parameters of SVM are influenced by the given feature subset. Both of the problems can be considered as combination optimization problems. Immune clonal algorithm offers natural and potential way to solve the task because of its characteristic of rapid convergence to global optimal solution. In the evolution, the suitable feature subset and optimal parameters are got simultaneously by minimizing the existing bound on the generalization error for SVM. The results of experiments on sonar data set show the effectiveness of the method.