hello

[Eric Y. Kow]

[Resending a third time due to error on my part...]

Hi Wren and Thomas,                                                                                                 
                                                                                                                    
Thanks for your messages!

Second, Thomas, sorry for not replying.  I didn't see your message and
consequently, Wren's reply because it got spam filtered!  I only noticed
it recently because I've been doing a little bit more recruiting lately,
hopefully these guys: http://homepages.cwi.nl/~jve/cs/ and Hal Daumé.
                                                                                                                    
I'm a research fellow at the University of Brighton.  I did my PhD on                                               
surface realisation, a subtask of natural language generation.  I plan                                              
on putting a generic chart parser some time as soon as I wring out the                                              
bugs.  It's got a toy CKY and Earley implementation, as well as a                                                   
reusable interactive debugger.                                                                                      
                                                                                                                    
Thanks, 

On Tue, Mar 24, 2009 at 19:47:02 -0400, wren ng thornton wrote:
> It seems <nlp at projects.haskell.org> is unroutable for the moment, but  
> here's my reply...

<snip unroutability details>>

> -------- Original Message --------
> Thomas Amundsen wrote:
> > I saw that the archive is empty. Does anyone post to this list?
> >
> > I am currently getting a Master's in Computational Linguistics, and I
> > am learning Haskell on my own, for fun.
> >
> > - Tom
>
> Hello Tom,
>
> So far, I think you're the first. Eric Kow started the list only
> recently (more for growing the nascent H+NLP community, rather than in
> response to an extant community, AFAIK). So, welcome!
>
> Introductions: My BA was in linguistics and anthropology, I just
> finished an MSE in computer science last fall (focusing on computational
> linguistics, artificial intelligence, and type theory), and next fall
> I'm starting a PhD program in cognitive science (focusing on the same).
> My general interests tend towards formal compositional systems and
> mathematical aspects of linguistics (e.g. morphosyntax, compositional
> semantics, grammar formalisms, automata theory, type theory, category
> theory...). I'm familiar with the statistical modeling side of things,
> though I'm really a linguist[1] at heart.
>
> If you're interested in statistical modeling or data mining, I have a
> couple of packages on Hackage which can be useful. Logfloat is for
> dealing with minuscule probabilities via the standard trick of
> projecting things into the log domain (though done prettily as a Num
> instance in Haskell :). And bytestring-trie is for using patricia trees
> to map strings to whatever (which can be paired with suffix arrays for
> doing sistring lookup).
>
> On the type theory side of things, (I haven't published any code for it
> yet, but) my MSE thesis was designing algorithms for many-sorted
> unification with recursive and non-linear types. I hope to have a
> simplified version written up for one of next year's conferences.
>
> So that's me, in one perspective on a large nutshell.
>
>
> [1] In the broadest sense of the term. Type theory, systems theory, and
> anthropology don't seem so different to me.
>
> -- 
> Live well,
> ~wren

-- 
Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow>
PGP Key ID: 08AC04F9
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 194 bytes
Desc: not available
Url : http://projects.haskell.org/pipermail/nlp/attachments/20090330/ca990d92/attachment.pgp