Thanks for pointing this out. Currently we do not distribute any models here at Apache, because we need to clarify if it is allowed to publish them under AL 2.0, which for example allows commercial use.
Our current models are still hosted at SourceForge. OpenNLP also includes converters for some corpora, we should check if we can add new converters for some of the corpora there easily. Jörn On 7/2/11 4:02 PM, Rick Kellogg wrote:
Greetings, I have found a great source of Corpus data at the following link: http://nltk.googlecode.com/svn/trunk/nltk_data/index.xml We might be able to use some of this data to generate models for distribution with OpenNLP. Just a thought. Rick
