I can't find any other references to this on the Web, and I don't have enough reputation to comment, so I'm writing a comment here about Alex MDRs answer.

Thankyou and I updated the code with your suggested change. A couple of questions:

1. weightsFeature (and weightsItem) can not be used when clusterAlg = "hc"?
2. If you use weightsFeature then shouldn't you set pFeature to something like 0.8. In my case I'm reducing the probability of selecting some features, and so they need to be excluded sometime, which only occurs when pFeature is less than 1 correct?