Effects of Empty Categories on Machine Translation

摘要: We examine effects that empty categories have on machine translation. Empty are elements in parse trees lack corresponding overt surface forms (words) such as dropped pronouns and markers for control constructions. start by training translation systems with manually inserted elements. find inclusion of some data improves the result. expand experiment automatically inserting these into a larger set using various methods modified corpus. show even when automatic prediction null is not highly accurate, it nevertheless end