> I programmed a hierarchical and a partioning Clustering based on the > lucene API. the lucene API offers some great methods, which are very > useful for clustering. if you want to programm your own solution: look for > scatter-gather, especially for grouper, which seems the most elaborated > solution: the STC clustering, based on phrases, incremental and fast.
You may also want to check the SVD-based algorithm we've developed called Lingo: http://citeseer.ist.psu.edu/631937.html http://portal.acm.org/citation.cfm?id=1069594.1069727&coll=GUIDE&dl=GUIDE&CFID=48792235&CFTOKEN=73030559 Here's a paper on how Lingo performs compared to e.g. STC: http://citeseer.ist.psu.edu/osinski04conceptual.html Cheers, Staszek -- Stanislaw Osinski, [EMAIL PROTECTED] http://www.carrot-search.com --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
