On 10/10/2013 12:54 PM, Jim wrote:
As far as I understood, the name finder is at the moment only stable
for one property, like person names. I would like to have the
traditional divison into persons, locations, organizations and misc.
When creating manually the training data, would it be OK to add all
four kinds already to the text and then, maybe create later 4 models
for the different properties?
There is no reason to create 4 different models. Just put all kinds of
NEs in the training set and the resulting model will be bale to
recognise all of them (assuming you've got enough data of course).
The documentation which marks it as experimental was written by me, I
did a couple of tests and it showed worse results than running them
individually.
Maybe that is not true for all data sets, I suggest to make your own
test. The built in evaluation can give you the performance numbers for
your data.
HTH,
Jörn
