Either would work. Spilling vectors from a Lucene index is more developed at this point in time and thus would probably be easier.
On Fri, Jan 1, 2010 at 7:45 AM, Bogdan Vatkov <[email protected]>wrote: > Do you mean I have to apply stemming during the vector creation or already > in Lucene indexing? Maybe from clustering POV it is the same but what would > you recommend? > -- Ted Dunning, CTO DeepDyve
