Predicate-Argument Structure-based Preordering for Japanese-English Statistical Machine Translation of Scientific Papers

Translating Japanese to English is difficult because they belong to different language families. Naı̈ve phrase-based statistical machine translation (SMT) often fails to address syntactic difference between Japanese and English. Preordering methods are one of the simple but effective approaches that can model reordering in a long distance, which is crucial in translating Japanese and English. Thus, we apply a predicate-argument structure-based preordering method to the Japanese-English statistical machine translation task of scientific papers. Our method is based on the method described in (Hoshino et al., 2013), and extends their rules to handle abbreviation and passivization frequently found in scientific papers. Experimental results show that our proposed method improves performance of both (Hoshino et al., 2013)’s system and our phrase-based SMT baseline without preordering.