data fusion similarity

Hi,

I am a newbie of pipeline pilot. I would like to do similarity search using 10 references against a database. I came across a paper: "Hassan et al., Cheminformatics analysis and learning in a data pipelining environment. Mol. Div. 2006, 283-299" were they used data fusion calculations for similar work.

Can someone give me an example protocol for "data fusion similarity"

It will also be helpful if you can give a sample script on how to calculate group similarity metric

Thanks
Diwan