Chinese-to-Japanese Patent Machine Translation based on Syntactic Pre-ordering for WAT 2015

摘要

This paper presents our Chinese-to-Japanese patent machine translation system for WAT 2015 (Group ID: ntt) that uses syntactic pre-ordering over Chinese dependency structures. A head word and its modifier words are reordered by hand-written rules or a learning-to-rank model. Our system outperforms baseline phrase-based machine translations and competes with baseline tree-to-string machine translations.