RE: [Corpora-List] wordlist-similarity tools in Java?

From: Nuno Seco (nseco@dei.uc.pt)
Date: Sun Feb 25 2007 - 16:05:25 MET

  • Next message: Megerdoomian, Karine: "[Corpora-List] CAASL-2: Last CFP and Deadline Extension"

    Hi.

    Well the library was intended to be used with nouns where the hyeronymyy
    backbone is strongest. Because I use the hyperonymy tree, in the same manner
    as Resnik, I'll always end up comparing synsets of the same POS.

    Nonetheless, I have also put up a Lucene index of wordnet (at the princeton
    site) where you can use that information if you wish. All you have to do is
    implement the similarity metric using the information you retrieve from the
    index.

    Cheers, 

    --
Nuno Seco

    > -----Original Message-----
> From: owner-corpora@lists.uib.no 
> [mailto:owner-corpora@lists.uib.no] On Behalf Of Andy Roberts
> Sent: sábado, 24 de Fevereiro de 2007 15:25
> To: Nuno Seco
> Cc: 'Eric Atwell'; CORPORA@UIB.NO; 'Stella Kleanthous'
> Subject: RE: [Corpora-List] wordlist-similarity tools in Java? 
> 
> Nuno,
> 
> That's a very handy library. However, may I ask why you don't 
> use POS information in your index (and therefore your 
> similarity algorithms)?
> Your library permits comparisons at a sense level but I don't 
> see how you compare sense 1 of live with dwell with out first 
> telling the program that you're only interested in the verb 
> and not the other two homographs.
> 
> Regards,
> Andy
> 
> On Fri, 23 Feb 2007, Nuno Seco wrote:
> 
> > Take a look at:
> > http://wordnet.princeton.edu/links#extensions
> >
> > You should find a library computing similarity using  Information 
> > Theoretic models in both Java and Prolog.
> >
> > Cheers,
> >
> > --
> > Nuno Seco
> >
> >> -----Original Message-----
> >> From: owner-corpora@lists.uib.no
> >> [mailto:owner-corpora@lists.uib.no] On Behalf Of Eric Atwell
> >> Sent: sexta-feira, 23 de Fevereiro de 2007 13:53
> >> To: CORPORA@UIB.NO
> >> Cc: Stella Kleanthous
> >> Subject: [Corpora-List] wordlist-similarity tools in Java?
> >>
> >>
> >> Stella Kleanthous, Leeds PhD student, asked me for a Java 
> >> tool/resource to measure the semantic similarity between 
> two lists of 
> >> (English) words.
> >> I directed her at WordNet-Similarity:
> >>
> >> http://www.d.umn.edu/~tpederse/similarity.html
> >> http://sourceforge.net/projects/wn-similarity
> >> http://search.cpan.org/dist/WordNet-Similarity/
> >>
> >> BUT this is Perl software;  has anyone implemented similar in Java?
> >> (which Stella (and others) could integrate into Java programs?)
> >>
> >> thanks
> >>
> >>
> >> Eric Atwell,
> >> Senior Lecturer, Language research group leader, School of 
> Computing, 
> >> Faculty of Engineering, University of Leeds, LEEDS
> >> LS2 9JT, England
> >> TEL: 0113-3435430  FAX: 0113-3435468  WWW: just Google eric atwell
> >>
> >
> >

    This archive was generated by hypermail 2b29 : Sun Feb 25 2007 - 16:04:10 MET