Hi, I'm building an Arabic-English SMT using Moses. As Arabic is a rich morphology language, I decided to use a morpholigically analyzed corpus to improve the quality of translations. As I know, to develope the corpus I need an English morphology analyzer and an Arabic one, that give a word (e.g: books in English) and gets the morphemes(book + s). Using these two analyzers I have to make a morphology analyzed corpus. Is it right? If yes, is there any English morphology analyzer which can do this?
I tried many tools such as Porter, freeling, ... but these tools remove the affixes and only return the stem of the word.
_______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
