Corpora: in-line PoS tagger

[Matthew Purver]

Thanks to everyone who helped me with this query.

Quite a few people suggested I write a wrapper script for the Brill tagger: this
is what I was doing before, and it certainly works - but it's a bit slow for my
purposes, so I was hoping to come across a faster alternative. Speed is
important as I'm intending to use this to manipulate user input (according to
PoS) in a "on-line chat"-type experiment (part of some ongoing work on
clarifications in dialogue).

Anyway, for those interested, here's the suggestions I received. At the moment
I'm using LTPOS because it's very easy to set up in a client-server
configuration, which (although I hadn't realised it before!) is exactly what I
needed:


Andrew Bredenkamp, Wojciech Skut and Chris Brew all suggested Thorsten Brants'
TnT (as did Thorsten himself):
http://www.coli.uni-sb.de/~thorsten/tnt/

Miles Osborne and Keith Miller suggested Adwait Ratnaparkhi's MXPOST:
ftp://ftp.cis.upenn.edu/pub/adwait/jmx/

Arno Erpenbeck suggested the IMS TreeTagger:
http://www.ims.uni-stuttgart.de/projekte/corplex/TreeTagger/DecisionTreeTagger.html

Oliver Mason suggested QTAG:
http://www.english.bham.ac.uk/staff/oliver/software/tagger/

Matsasugu Ono suggested NYU's Apple Pie Parser:
http://cs.nyu.edu./cs/projects/proteus/app/

Roxana Angheluta suggested Edinburgh's LTPOS:
http://www.ltg.ed.ac.uk/software/

Ingo Schroeder suggested ICOPOST:
http://nats-www.informatik.uni-hamburg.de/~ingo/icopost/


thanks to all,
Matt

--
Matthew Purver  <matthew.purver at kcl.ac.uk>

Computational Linguistics and Natural Language Processing Group
Department of Computer Science
King's College London, Strand, London WC2R 2LS