Unlike the other openNLP components you only need a couple of dozen
sentences for an adequate sentence-detection model. You can easily find
50-60 sentences can't you?
Jim
On 08/01/13 22:44, Jean-Philippe Fauconnier wrote:
Hello,
Annotate (a lot of) data is a prerequisite. See methodology
of corpus building in the literature.
After that, either use directly MaxEnt OpenNLP, or use Sentence
Detection API.
See on
http://opennlp.apache.org/documentation/1.5.2-incubating/manual/opennlp.html#tools.sentdetect.detection.api
Regards
J. Fauconnier
Le 08/01/2013 20:17, ali koyuncu a écrit :
Hi, I want to create a sentence model for the Turkish Language.
Anyone have
advice on practical ways to do this?
