Hi, I am trying to use svmB in MLInterfaces for my analysis, and results look very different from what I got using svm in GeneSpring. Does anyone here have experience with MLInterfaces, especially svmB? how the implementation of svm in MLInterfaces (or R) differ from that in GeneSpring? Thanks a lot.

Hi I don't know much about the Genespring software but you may want to compare the various SVM control parameters. Or have you already done this? The important ones are usually kernel, degree, gamma, and cost you can see these parameters in the function typing >svmB with no call I think that the MLInterfaces method is a wrapper (i.e. a consistent interface) for the svm function in package "e1071". The documentation for this package is fairly sparse as you are expected to look at the originating methods for details. so try: >library(e1071) >?svm to see specific details of these parameters Stephen Henderson Wolfson Inst. for Biomedical Research Cruciform Bldg., Gower Street University College London United Kingdom, WC1E 6BT +44 (0)207 679 6827 -----Original Message----- From: bioconductor-bounces@stat.math.ethz.ch [mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of Qi.Zhang at astrazeneca.com Sent: 06 March 2006 22:35 To: bioconductor at stat.math.ethz.ch Subject: [BioC] MLInterfaces Hi, I am trying to use svmB in MLInterfaces for my analysis, and results look very different from what I got using svm in GeneSpring. Does anyone here have experience with MLInterfaces, especially svmB? how the implementation of svm in MLInterfaces (or R) differ from that in GeneSpring? Thanks a lot. _______________________________________________ Bioconductor mailing list Bioconductor at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/bioconductor ********************************************************************** This email and any files transmitted with it are confidentia...{{dropped}}