You are viewing a plain text version of this content. The canonical link for it is here.

Posted to users@opennlp.apache.org by Tim Schmolka <ti...@schmolka.org> on 2015/11/11 13:53:28 UTC

Spanish Models

Hey Guys,

just a quick question:
Does OpenNLP support spanish models? (Things such as tokenization, sentence
detection, etc.) ?

Best regards,
Tim

Re: Spanish Models

Posted by Rodrigo Agerri <ra...@apache.org>.

Hello,

There are some models for Named Entity Recognition in

http://opennlp.sourceforge.net/models-1.5/

However, it is possible to train your own models with the Ancora corpus

http://clic.ub.edu/corpus/ancora

which is free and includes annotation for almost every component of
OpenNLP (POS, NER, Parsing, etc.).

For the record, other tools that provide trained Spanish models are:
Freeling, Stanford CoreNLP, Mate tools (GPL licensed) and IXA pipes
(Apache 2.0 License and based on OpenNLP Machine Learning API).

HTH,

R

And the parser supports training

On Wed, Nov 11, 2015 at 1:53 PM, Tim Schmolka <ti...@schmolka.org> wrote:
> Hey Guys,
>
> just a quick question:
> Does OpenNLP support spanish models? (Things such as tokenization, sentence
> detection, etc.) ?
>
> Best regards,
> Tim