>
> Maybe somebody can help me.
> I'm working on a project to extract bilingual lexicons from pre-aligned
> parallel corpora.
> Now I found some information about related projects but somtimes
> this information is quite short and more detailed information is hard to
> obtain. Especially some articles in proceedings of ACL workshops or Annual
> Meetings sound quite interesting to me but I couldn't find them or order
> them in the libraries I can use.
> So maybe it is possible to get some of them in an electronical saved
> version or somehow else.
>
> It would also be interesting to see some test versions or some test
> results of systems like
> Champollion (Smadja...)
> char_align (Church)
> Candide (Brown)
> Termight (Dagan/Church)
> Bicord (Klavans/Tzoukermann)
> or other approaches (Kupiec, van der Eijk, Fung/McKeown, Wu/Xia...)
>
One generally good source for paper in computational linguistics is
"The Computation and E-print Archive":
http://xxx.lanl.gov/cmp-lg/
Some (admittedly not very many) of the papers you mention above can be
found there.
You can also find several of our own papers about parallel texts on
our site:
http://casoar.citi.doc.ca:4321
try an on-line application of parallel texts (a bilingual
concordancer), follow the "TransSearch" link.
-- Pierre
-- Pierre Isabelle tel: (514) 973-5801 CITI, 1575 Chomedey Blvd, fax: (514) 973-5757 Laval, Quebec, Canada H7V 2X2 e-mail: isabelle@citi.doc.ca