[Corpora-List] free parser for english

Jason Baldridge jasonbaldridge at gmail.com
Tue Jan 16 15:46:07 UTC 2007


Here are three Java parsers that may be of use:

Dan Bikel's: http://www.cis.upenn.edu/~dbikel/software.html#stat-parser<http://www.cis.upenn.edu/%7Edbikel/software.html#stat-parser>
OpenNLP Parser: https://sourceforge.net/projects/opennlp/
MSTParser: http://www.seas.upenn.edu/~strctlrn/MSTParser/MSTParser.html
<http://www.seas.upenn.edu/%7Estrctlrn/MSTParser/MSTParser.html>

You may need to train models for English (though I believe there is an
English parsing model available for the OpenNLP parser, which is based on
Ratnarparkhi's maxent parser). If you do need to train models, small amounts
of the Penn Treebank are available as part of the NLTK Toolkit, but for more
accurate models you'll need access to the full Penn Treebank.

You might also want to check out the C&C CCG parser, which comes with models
and outputs various types of dependency structures:

http://svn.ask.it.usyd.edu.au/trac/candc/wiki

You won't get any tree structures from this parser (it produces categorial
grammar derivations), but probably the dependency relations are all you need
anyway. It is also one of the fastest parsers on offer. (But note that free
usage is only for academic purposes.)

Jason

On 1/16/07, ben dbabis samira <bendbabis_samira at yahoo.fr> wrote:
>
> hi,
> Thanks for answering to my mail.
> I need exactly a free parser for english in java that generates as
> output syntactic tree (with semantic dependency relations) of the sentence
> introduced as input.
> best regards;
> Samira BEN DBABIS
> MIRACL laboratory
> Sfax, TUNISIA
>
> ------------------------------
> Découvrez une nouvelle façon d'obtenir des réponses à toutes vos questions
> ! Profitez des connaissances, des opinions et des expériences des
> internautes sur Yahoo! Questions/Réponses<http://fr.rd.yahoo.com/evt=42054/*http://fr.answers.yahoo.com>
> .
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20070116/e22105d3/attachment.htm>


More information about the Corpora mailing list