-----BEGIN PGP SIGNED MESSAGE-----
I am looking for corpora of tagged Arabic texts for an application
dealing with automated grammar extraction and reference grammar
generation using neural networks.
Because we have yet to determine whether this particular application
of neural networks is technically possible, it would be of advantage
if the texts' grammar was not overly complex.
On the other hand, since the underlying principles are quite similar
other Semitic languages, a corpus of tagged Hebrew, Akkadian or Geez
would probably do as well for the evaluation of possibility (The
actual application will use Qur'anic Arabic)
I'd be very grateful for information of any kind in this respect.
The Tao that is seen / Is not the true Tao, until / You bring fresh
-----BEGIN PGP SIGNATURE-----
Version: PGP 6.5i
Comment: Freedom of the press is limited to those who own one.
-----END PGP SIGNATURE-----
This archive was generated by hypermail 2b29 : Thu Sep 14 2000 - 12:24:31 MET DST