Hi, you could learn a statistical post-editing model: Train a statistical machine translation model with the rule-based MT's translation as input and the real translation as output.
-phi On Tue, Jun 29, 2010 at 7:45 AM, 亮 田 <[email protected]> wrote: > Dear sir, > Recently ,I use the rule-based machine translation to get the translation > results. But the result is not very fluent. Now I want to use the Moses > decoder to optimize the result,Can you kindly enough to give me some > suggestion? > For example: > one result is : Boys sleep in a dormitory who > live at school. > Reorder the result to :Boys who live at school sleep in a > dormitory. > Thank you very much for your reply! > > Liang > > > _______________________________________________ > 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
