Improving Sequence to Sequence Neural Machine Translation by Utilizing Syntactic Dependency Information

Sequence to Sequence Neural Machine Translation has achieved significant performance in recent years. Yet, there are some existing issues that Neural Machine Translation still does not solve completely. Two of them are translation of long sentences and “over-translation”. To address these two problems, we propose an approach that utilize more grammatical information such as syntactic dependencies, so that the output can be generated based on more abundant information. In addition, the output of the model is presented not as a simple sequence of tokens but as a linearized tree construction. Experiments on the Europarl-v7 dataset of French-toEnglish translation demonstrate that our proposed method improves BLEU scores by 1.57 and 2.40 on datasets consisting of sentences with up to 50 and 80 tokens, respectively. Furthermore, the proposed method also solved the two existing problems, ineffective translation of long sentences and over-translation in Neural Machine Translation.