I'm currently reasonably new to R and am having trouble extracting the information I would like from a package.
I am using
MLSeq to implement Random Forest on RNA Seq data to find biomarkers for a condition. Currently, the output given by default is just how well it classified the data and a table that describes actual class against predicted class.
What I want is the importance of each feature so that i can take the highest ranking features and continue to investigate those.
Does anyone have experience with MLSeq package or know of a similar machine learning package that has this functionality?
Best How To :
the caret package has a very useful function called varImp (http://www.inside-r.org/packages/cran/caret/docs/varImp). If you don't have a very large number of predictors you could use it to get/plot their importance. In your case, let's suppose you have trained your model:
svm = classify(data = data.trainS4, method = "svm", normalize = "deseq", deseqTransform = "voom", cv = 3, rpt = 3, ref = "PP")
you can get the variable importance of your predictors using the following command:
VI <- varImp([email protected]).
However, before doing that, read carefully how varImp() works