2013/1/31 Gael Varoquaux <gael.varoqu...@normalesup.org>: > On Thu, Jan 31, 2013 at 08:17:02AM -0500, Satrajit Ghosh wrote: >> the problem that this causes is in comparing clusters from two similar >> datasets. are the clusters different because they reflect different >> properties or simply differences in local minima. > > Because clustering is so much of an ill-posed problem in noisy and > high-dimensional situations, as in neuroimaging, such comparison is > difficult and often imposible. > > Quite often people believe way too much in the clustering results that > they get: it is not clear that the clusters are identifiable from the > data.
Ensemble clustering might help you identify stable substructures (e.g. a ranking on the stability of pairs of samples for instance). -- Olivier http://twitter.com/ogrisel - http://github.com/ogrisel ------------------------------------------------------------------------------ Everyone hates slow websites. So do we. Make your web apps faster with AppDynamics Download AppDynamics Lite for free today: http://p.sf.net/sfu/appdyn_d2d_jan _______________________________________________ Scikit-learn-general mailing list Scikit-learn-general@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
