Hello
it is there to measure the distribution of the outcome.
E.g. for the name it could be:
start 5%
cont 10%
other 85%
If you remove it in the name finder you get a slightly higher recall,
and bit lower precision.
HTH,
Jörn
On 10/11/2012 06:42 PM, Jim foo.bar wrote:
Hi guys,
I' can't figure out for the life of me what this feature-generator
does! I originally thought it was generating features based on the
outcome of a previous token but that is what
PreviousMapFeatureGenerator does... The docs state:
"The definition feature maps the underlying distribution of outcomes."
What does that mean? This description is somewhat different from all
the rest...
thanks in advance,
Jim
