Hi,

for Arabic, a commonly used pre-processing suite is MADA and TOKAN,
which you can get here:
http://www1.ccls.columbia.edu/MADA/index.html

-phi

On Mon, Mar 25, 2013 at 8:20 PM, Mustafa Helal <[email protected]> wrote:
> Hello
> i just finished instillation and training moses system with french data; but
> i need to test it using arabic data
> my problem i don't now how to proceed with "tokenizer" step on ARABIC case
> Also what should take care of while doing such training? may be something
> like encoding
>
> --
> Regards,
> Mustafa Helal
>
>
> _______________________________________________
> Moses-support mailing list
> [email protected]
> http://mailman.mit.edu/mailman/listinfo/moses-support
>
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support

Reply via email to