User talk:Karthikprasad/GSOC 2012 proposal
Hi Karthikprasad 
Some ideas. I recently found a paper about similar work done by linguists. I will try to find it and contact them and see if they are interested in contributing their code and co-operating with us. They created an unsupervised POS taggers which uses only 7 POS tags. The approach describes seems to work across many languages.
See Also 
- http://nlp.cs.nyu.edu/wikipedia-data/ for a good overview and some suggestion on tools you could use.
- I am so sorry! I somehow missed looking into this page! Pardon me for the terribly late reply. Thank you so much for the heads-up. I see that a little bit of what our project intends to do has been done. But there still is a lot to do. This can, however, help us to step up our work and enlarge the scope our project. Karthik