On Tue, Mar 07, 2017 at 03:54:25PM +0530, Ayush Pandey
wrote:> Wanted to know if other normalization techniques like
> normalization by standard-deviation have been tried to normalize the
> Feature-list in Letor.
I asked Parth about this, and his reply was:
basically we don't have tried different normalisation methodologies but
what we have in place is QuerylevelNorm which is a slight variant of
divide by maximum value limited to that query. Link is here:
https://trac.xapian.org/wiki/GSoC2011/LTR/Notes#QueryLevelNorm
Cheers,
Olly