Hi Wenlong The moses web translation service organises the tokenisation/de-tokenisation etc for you by calling the moses scripts. The moses engine is only intended to do translation, with tokenising and casing issues deferred to separate scripts,
regards Barry On Monday 23 Aug 2010 07:48:33 you wrote: > Hi All, > > We made a web service based on the information on this page: > http://www.statmt.org/moses/?n=Moses.WebTranslation > > > The web service works, but I can see some flaws in it: > It only calls the Moses engine itself to do the translation in the > daemon.plscript, but during our evaluation steps, there are extra > scripts to perform > the "tokenizer", "recaser', "detokenizer" and etc need to be run to make > the quality to be good. > > So is there any implementations already there whcih helps to add above pre > and post processings before adn after the Moses transaltion? > Or we have to perfect this ourselves? If there are already such kind of > resource, that would be cool~ ;-) > > > Thanks so, > Wenlong > _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
