GLRT和LS_SVM应用于基因表达数据分类

为快速、准确地对基因芯片表达数据进行分类,提出了一种新型的基因芯片表达数据分类模型.该模型首先使用广义似然比检验(GLRT)有效鉴别出表达有显著性差异的基因.然后,将这些表达有显著性差异的基因用于最小二乘支持向量机(LS_SVM)的训练,从而建立了基于GLRT+LS_SVM的基因芯片表达数据分类模型.该模型在处理数据量大、维数高、样本量小、非线性等特点的基因芯片数据时有很大优势,可以广泛用于处理基因芯片数据.

Generalized Likelihood Ratio Test and Least Square Support Vector Machine Applied to the Classification of Microarray Gene Expression Data

A new analysis model combined Generalized Likelihood Ratio Test(GLRT) with Least Square-Support Vector Machine(LS_SVM) is developed to perform fast and accurate classification for microarray expression data.In this model,GLRT is firstly adopted to find differentially displayed genes at the expression level,and then these found genes are trained with LS_SVM,namely GLRT+LS_SVM microarray expression data classification model.The results demonstrate the feasibility of this model in solving the problem of classification in the nonlinear,high dimension and limited-case samples,suggesting an effective strategy for discovering and predicting cancer classes.