You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@opennlp.apache.org by Ant B <am...@gmail.com> on 2012/12/12 01:29:46 UTC

DefaultLinker, chunker parser and NER annotation

Dear OpenNLP community,

I am currently working with OpenNLP in order to explore entity extraction
as a development project.

I have successfully used the sentence detector, word tokenier, POS tagger
and the pre-defined named entity models.

In addition, having read the following:

I have managed to get an example of the coreference tool running.

While the results on test sentences seem sensible, I am not confident that
my code is correct.  The two main reasons being:

1.  The implementation details on the blog site are taken from the
TreebankLinker class;

2.  I do not passing NER annotations to the DefaultLinker within the
sentence Parse objects.

Efforts to date:

following this post:

I inspected the TreeBankNameFinder class, but could not see how to apply
this string building process to create Parse objects.



With thanks,


Ant