yes there is a solr component if you want to cluster solr documents , check the following link http://wiki.apache.org/solr/ClusteringComponent
Carrot2 might be good if you want to cluster few thousands of documents , for example when user search solr , just cluster the search results Mahout is much more scalable and probably you need Hadoop for that thanks chandan On Tue, Sep 4, 2012 at 2:10 PM, Denis Kuzmenok <forward...@ukr.net> wrote: > > > -------- Original Message -------- > Subject: Solr Clustering > From: Denis Kuzmenok <forward...@ukr.net> > To: solr-user@lucene.apache.org > CC: > > Hi, all. > I know there is carrot2 and mahout for clustering. I want to implement > such thing: > I fetch documents and want to group them into clusters when they are added > to index (i want to filter "similar" documents for example for 1 week). i > need these documents quickly, so i cant rely on some postponed > calculations. Each document should have assigned cluster id (like group > similar documents into clusters and assign each document its cluster id. > It's something similar to news aggregators like google news. I dont need > to search for clusters with documents older than 1 week (for example). Each > document will have its unique id and saved into DB. But solr will have > cluster id field also. > Is it possible to implement this with solr/carrot/mahout? -- Chandan Tamrakar * *
