On 12/13/2012 04:08 PM, Steven Bethard wrote:
On Dec 13, 2012, at 4:04 PM, Jörn Kottmann <[email protected]> wrote:
On 12/13/2012 01:56 PM, Miller, Timothy wrote:
Any interest in generifying the relation extractor to make it easier to
instantiate all relation extractors as clearTK style annotators? This would
make it easier to share feature extractors, allow me to rewrite the coreference
code base, and make it easier to contribute to multiple projects without
learning a bunch of different implementations.
What do you think about starting a relation extractor component over at
OpenNLP?
I suspect this is probably unlikely, given that we're already using
ClearTK-style training, feature extraction, etc. and I'm pretty sure OpenNLP
doesn't want to depend on ClearTK modules...
The only requirement for me which is really important is that it is
generic and can be trained/adapted for the news domain,
usually I try to avoid the dependency on UIMA in my NLP components
because it makes them hard to integrate when you
do not want to use UIMA. UIMA is a nice tool when you need to process
large amounts of texts with a couple of NLP tools put
together in someway, but it makes things difficult if you try to do
something with just one text or a snippet of it (e.g. to produce
relations as input for an annotation tool).
Jörn
