Find Unbalanced Clusters


It seems Cluto always makes balanced clusters, i.e. the number of item in each cluster is at the same scale. However, this is not the case for my data, as well as for many other cases, I guess.

Is it possible for Cluto to find clusters as unbalanced as they naturally are? Or what modification can we do to achieve it?



RE: You can get unbalanced

You can get unbalanced clusters by using -clmethod=direct.