[developers] [itsdb] Problem with scores on my inflection rules

[Woodley Packard]

I took a look at the relevant ACE code and I believe that Stephan is correct:  MEM feature matching is case sensitive.  It will be interesting to follow the trail back to see how the feature in the model became all uppercase.

Woodley

On Dec 7, 2012, at 12:07 PM, Stephan Oepen wrote:

> the one surprising-looking thing in your MEM features is the upper
> case in the forms of lexicalized features, e.g. "TIL".  i would have to
> check the code, but one suspicion i have is that PET and ACE look
> up these features with the surface form either all downcased, or in
> exactly the variant used as the parser input (presumably not ‘TIL’ in
> all upper case, in your inputs, either way).

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.delph-in.net/archives/developers/attachments/20121207/bd45cb0b/attachment.html>