The Role of large Monolingual Corpora in Improving MachineTranslation Quality;

Collocations, recurrent combinations of words whose co-occurrence probability ishigher than chance, are frequent in natural languages. since bilingual dictionariesdo not offer proper equivalents for most of such collocations, the majority ofmachine translation systems perform poorly when faced with collocations, and asa result, their output quality decreases to a high degree. monolingual corporahave recently been used in solving various linguistic problems including naturallanguage processing, statistical machine translation, language teaching and thelike.the present study describes the process of creating and exploiting a largemonolingual corpus of persian. this corpus enables us to solve the ambiguityproblem of english collocations when translating into persian with an englishpersianmachine translation system. using such a corpus as a target languagecorpus and an english-persian bilingual dictionary, we study the efficiency of thiscorpus in finding the most appropriate persian equivalents for english collocationsin order to enhance the output quality of a machine translation system. the resultsof the experiment on a test corpus were very encouraging and achieved 90.83%success.