A Linguistically Annotated Reordering Model for BTG-based Statistical Machine Translation

In this paper, we propose a linguistically anno-tated reordering model for BTG-based statis-tical machine translation. The model incorpo-rates linguistic knowledge to predict orders for both syntactic and non-syntactic phrases. The linguistic knowledge is automatically learned from source-side parse trees through an an-notation algorithm. We empirically demon-strate that the proposed model leads to a sig-nificant improvement of 1.55 % in the BLEU score over the baseline reordering model on the NIST MT-05 Chinese-to-English transla-tion task