To solve the classification problem with the help of ranking the features an algorithm was proposed by Guyon, Isabelle, et al. named SVM-RFE. In this algorithm the dataset has been trained with SVM linear kernel model and the feature containing the smallest ranking is removed. This criterion is the w value of the decision hyperplane given by the SVM.
x n-by-d data matrix to train (n samples/patients, d clones/genes)
y vector of class labels -1 or 1\'s (for n chips/patients )
Adopted from R code: http://www.uccor.edu.ar/busquedas/?txt_palabra=seminarios
returns the feature list.
This function also rank the feature.
Guyon, Isabelle, et al.
Guyon, Isabelle, et al. "Gene selection for cancer classification using support vector machines." Machine learning 46.1-3 (2002): 389-422.
Zhang, H. H., Ahn, J., Lin, X. and Park, C. (2006). Gene selection using support vector machines with nonconvex penalty. Bioinformatics, 22, pp. 88-95.
scadsvc, predict.penSVM, sim.data
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
#Example for svmrfeFeatureRanking() #Data set taken from GSE2280 library(SummarizedExperiment) data(ExampleRawData, package="sigFeature") x <- t(assays(ExampleRawData)$counts) y <- colData(ExampleRawData)$sampleLabels x <- x[ ,1:500] #featureRankedList = svmrfeFeatureRanking(x,y) print(featureRankedList[1:10]) #Train the data with ranked frature #library(e1071) #svmmodel = svm(x[ , featureRankedList[1:50]], y, cost = 10, kernel="linear") #summary(svmmodel)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.