I've been working on similar stuff and have had a poster at last year's LREC: http://ufal.mff.cuni.cz:8080/bib/?section=publication&id=-6437616343801484763&mode=view (and the framework is here: https://wiki.ufal.ms.mff.cuni.cz/user:zeman:interset )
However, my universal tagset is a virtual one - it's a definition of possible features and their values, not exactly a set of tags (encoded as strings). Also, it's work in progress, and changes will be needed to achieve universality. Anyway, let me know if I can be of any help.
Best, Dan
Adam Teichert napsal(a):
> Hello all.
>
>
> I've been looking for a POS tagset that is general enough to
> effectively tag "any" natural language. (I'm looking at Linguistic
> Typology / Universal Implications so I want to compare POS taggings
> across many [possibly obscure] languages.) Does anyone know of such a
> tagset?
>
> If anyone is interested in what I've found so far, this paper seems relevant:
> "Induction of Fine-grained Part-of-speech Taggers via Classifier
> Combination and Crosslingual Projection" (Elliott Franco Dr´abek,
> David Yarowsky)
> http://acl.ldc.upenn.edu/W/W05/W05-0807.pdf
>
> Also, I'm aware of some efforts at Microsoft Research India, to
> perhaps develop a "universal" tagset for Indian Languages:
> http://research.microsoft.com/en-us/groups/mls/default.aspx
>
>
> Thanks for any ideas.
>
> --Adam (R. Teichert)
>
> MS Student
> School of Computing
> University of Utah
>
> _______________________________________________
> Corpora mailing list
> Corpora at uib.no
> http://mailman.uib.no/listinfo/corpora
>
-- RNDr. Daniel Zeman, Ph.D. ÚFAL MFF, Univerzita Karlova, Praha http://ufal.mff.cuni.cz/~zeman/