Gene selection through sensitivity analysis of support vector machines

Authors:
Defeng Wang;Daniel S. Yeung;Eric C. C. Tsang;Lin Shi
Affiliations:
Department of Computing, The Hong Kong Polytechnic University, Kowloon, Hong Kong, China;Department of Computing, The Hong Kong Polytechnic University, Kowloon, Hong Kong, China;Department of Computing, The Hong Kong Polytechnic University, Kowloon, Hong Kong, China;Department of Computer Science and Engineering, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong, China
Venue:
CompLife'05 Proceedings of the First international conference on Computational Life Sciences
Year:
2005

Citing 6
Cited 0

The nature of statistical learning theory

The nature of statistical learning theory
Selection of relevant features and examples in machine learning

Artificial Intelligence - Special issue on relevance
Choosing Multiple Parameters for Support Vector Machines

Machine Learning
Gene Selection for Cancer Classification using Support Vector Machines

Machine Learning
An introduction to variable and feature selection

The Journal of Machine Learning Research
Variable selection using svm based criteria

The Journal of Machine Learning Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a novel approach to gene selection for microarry data through the sensitivity analysis of support vector machines (SVMs). A new measurement (sensitivity) is defined to quantify the saliencies of individual features (genes) by analyzing the discriminative function in SVMs. Our feature selection strategy is first to select the features with higher sensitivities but meanwhile keep the remaining ones, and then refine the selected subset by tentatively substituting some part with fragments of the previously rejected features. The accuracy of our method is validated experimentally on the benchmark microarray datasets.