svm.fs
(Package: penalizedSVM) :
Fits SVM mit variable selection using penalties.
Fits SVM with variable selection (clone selection) using penalties SCAD, L1 norm, Elastic Net (L1 + L2 norms) and ELastic SCAD (SCAD + L1 norm). Additionally tuning parameter search is presented by two approcaches: fixed grid or interval search.
Simulation of 'n' samples. Each sample has 'sg' genes, only 'nsg' of them are called significant and have influence on class labels. All other '(ng - nsg)' genes are called ballanced. All gene ratios are drawn from a multivariate normal distribution. There is a posibility to create blocks of highly correlated genes.
Feature Selection SVM using penalty functions. The smoothly clipped absolute deviation (SCAD) and L1-norm penalties are availible up to now. Other functions will be implemented in the near feature.