Incorporating gene similarity into support vector machine for microarray classification and gene selection

摘要

In this paper, we propose a novel method based on support vector machine (SVM) for microarray classification and gene (feature) selection. The proposed method, called similaritybased SVM (SSVM), incorporates the prior knowledge of gene similarity into the standard SVM by combining the standard l2 norm and the similarity penalty of all the genes. The preliminary experiments show that our method performs better than the standard SVM, l2 l0 SVM and SVMRFE, especially when the features are highly similar.