Abstract
T-score between classes and gene expressions is widely used for gene ranking in microarray gene expression data analysis. We propose to use only support vector points for computation of t-scores for gene ranking. The proposed method uses backward elimination of features, similar to Support Vector Machine Recursive Feature Elimination (SVM-RFE) formulation, but achieves better results than SVM-RFE and t-score based feature selection on three benchmark cancer datasets. © 2008 Springer Berlin Heidelberg.
Cite
CITATION STYLE
Mundra, P. A., & Rajapakse, J. C. (2008). Support vector based T-score for gene ranking. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5265 LNBI, pp. 144–153). Springer Verlag. https://doi.org/10.1007/978-3-540-88436-1_13
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
