In this proposed work, word re-ordering rules for English toDravidian languagesMachine TranslationSystem is developed.Machine Translation (MT) mainly deals with the transformation of text from onenatural language toanother. Generally in Machine translation, the transformation hypothesis iscomputationally expensive. If we perform arbitrary reordering of words while translation, the searchproblem will be NP-hard.If we restrict the possible reordering in an appropriate way, we obtain apolynomial-time search algorithm.Word alignment differs for a sentence from one language to anotherlanguage while perform machine translation. Re-ordering the given sentence improves the performance ofthe statistical translation. Here word re-ordering plays a very important role. In this paper, wordreordering system for English toDravidian languagesis made tosolve this problem in machinetranslation. The dependency information's are retrieved from Stanford parser incorporating with it; rulesare created for the transformations. Transformations are then applied toEnglish language to obtain theword order closer to thetargetDravidianlanguages.This word reordering approach ispreprocessing toolfor machine translationwhich significantly improves the machine translation
展开▼
