I've been digging into the Mahout documentation It seems like LDA may work
Has anyone considered using LDA for directed graph clustering? On Sat, Jul 2, 2011 at 6:41 PM, Timmy Wilson <[email protected]> wrote: > Hi mahout community, > > I'm interested in using the following method to cluster a large > directed social graph > > Overlapping Community Detection using Bayesian Nonnegative Matrix > Factorization - > http://www.oxford-man.ox.ac.uk/documents/papers/2011OMI07_Roberts_et_al.pdf > > Can this be done in mahout? >
