Thanks for the reply, Tommaso! I would like to see tighter integration like in the way Nutch integrates with Solr. There is a single param that you set which points to the Solr instance. My interest in Mahout is with it's abitlity to handle large data and find frequency, co-location of data, clustering, etc...All the algorithms that are in the core build are great and I am just now wrapping my head around how to use them all.
Adam On Thu, Jun 9, 2011 at 10:33 AM, Tommaso Teofili <tommaso.teof...@gmail.com>wrote: > Hello Adam, > I've managed to create a small POC of integrating Mahout with Solr for a > clustering task, do you want to use it for clustering only or possibly for > other purposes/algorithms? > More generally speaking, I think it'd be nice if Solr could be extended > with > a proper API for integrating clustering engines in it so that one can plug > and exchange engines flawlessly (just need an Adapter). > Regards, > Tommaso > > 2011/6/9 Adam Estrada <estrada.adam.gro...@gmail.com> > > > Has anyone integrated Mahout with Solr? I know that Carrot2 is part of > the > > core build but the docs say that it's not very good for very large > indexes. > > Anyone have thoughts on this? > > > > Thanks, > > Adam > > >
