[Corpora-List] Train corpora for Named Entity Recognition

[Sundeep Raikhelkar]

Hi,

I have downloaded NLTK recently. I used nltk.ne_chunk() for Named Entity
Recognition using maxent multi-class chunker. I am amazed at the output. But
I still need to train this further to improve accuracy and more to
add domain specific labels.

I am looking for tutorial / manual to train this multi-class chunker (is
possible using an example).
Could anybody please let me know if there is any material / book online for
this.

Thanks,
Sundeep
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/corpora/attachments/20091111/2d69262a/attachment.htm>
-------------- next part --------------
_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora