Why ignore them? They seem to be pretty awesomely clustered! One reasonable approach is to remove duplicate items before clustering.
On Wed, Sep 29, 2010 at 10:45 AM, Derek O'Callaghan <[email protected] > wrote: > I'm wondering if the quickest and easiest solution is to simply ignore such > clusters [with duplicate members], i.e. those that currently generate a NaN > std? I'm not sure if it's the "correct" approach though... >
