On Wednesday, September 19, 2012 12:40:00 AM UTC+5:30, Mark Lawrence wrote: > On 18/09/2012 19:35, subhabangal...@gmail.com wrote: > > > Dear Group, > > > If anyone of the learned members can kindly help with a HMM/CRF based > > chunker on NLTK. > > > > > > Regards, > > > Subhabrata. > > > > > > > Certainly but how do you intend paying us? :) > > > > An alternative approach is to provide us with an idea of what you've > > researched, what code you've written and what problems you've got with > > it. If you get a code exception please provide the complete traceback > > and a mimimal code snippet that reproduces the issue. This way you're > > much more likely to get help. > > > > -- > > Cheers. > > > > Mark Lawrence.
Two possibilities Mark, either nltk.chunk.(something) or nltk.tag.hmm etc. I am thinking to experiment Regular Expressions, Unigram/Bigram based chunkers are running good even with Indian languages. HMM tag is also going great. I am thinking on the issue. Regards, Subhabrata. -- http://mail.python.org/mailman/listinfo/python-list
