If you have only probabilistic models, you can use too a stacking
strategy, combining the outputs of yours multiple models in a simple
logistic regression. And why not, add some extra features.
J.
On 13/02/14 10:52, Jörn Kottmann wrote:
On 02/13/2014 08:42 AM, Amir Pouya Agha Sadeghi wrote:
Hello Guys,
Ok, I changed the feature generation file and improved my result.Now
I want
to combine my open nlp name finder result with my Stanford NER
result. My
first idea was a simple voting system but since I have only two
system and
it can be a tie, I can not do that. Any one ever did that sort of things
before?
You could use more models, in OpenNLP you can train a perceptron and
maxent model.
The perceptron model might work much better, depending on your data.
In the latest trunk version we now also have support for mallet maxent
and crf.
HTH,
Jörn
