Hi, thanks for great update, rest is inlined:

On Sun, Jul 21, 2019 at 09:50:52PM +0000, Amr Mohamed Hosny Anwar wrote:


> Looking forward to reading your advice on how to proceed with the project.
> Additionally, Do you think we can make use of a parallel corpus for two 
> languages in some way or another?
> I know a parallel corpus is also somehow supervised but my intuition is 
> that finding/developing parallel corpora is easier than 
> finding/developing a tagged corpus.

This idea is interesting, I think it would fit well as an extension to
the experiment where you use counts from undisambiguated analyses as
weights or so. As I see it, it kind of seeks evidence of analyses based
on assumption that in translated sentence the tags of correct
translation/analysis would be the same, the extent to how much this
holds true is somewhat correlated ot the distance of the language pair
in typological terms, and the kind of sentences you have as well, the
genre and type of corpus, like a parallel corpus that's proper
translationese will surely work with this approach whereas probably
profesionally translated literature the less so. In any case, if you can
sketch and build an experiment in few days it's worth looking into. 


-- 
Doktor Tommi A Pirinen, Computational Linguist,
<https://flammie.github.io/purplemonkeydishwasher/>, Universität
Hamburg, Hamburger Zentrum für Sprachkorpora <http://hzsk.de>. CLARIN-D
Entwickler.  President of ACL SIGUR SIG for Uralic languages
<http://gtweb.uit.no/sigur/>.
I tend to follow inline-posting style in desktop e-mail messages.

Attachment: signature.asc
Description: PGP signature

_______________________________________________
Apertium-stuff mailing list
Apertium-stuff@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/apertium-stuff

Reply via email to