[Corpora-List] Document similarity tools?

[Torsten Zesch]

Hi Ivelina,

> I was wondering whether there is a public library or toolbox including various
> document similarity measures.

DKPro Similarity
http://code.google.com/p/dkpro-similarity-asl/
is a pretty comprehensive framework including most of the well-known measures.

We also successfully used it in our submission for last year's SemEval "Semantic Text Similarity" shared task:
http://aclweb.org/anthology-new/S/S12/S12-1059.pdf

-Torsten

_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora