基于神经网络的英文机辅翻译预调序模型研究

Research on neural network based preorder model of English machine-aided translation

长距离调序由于缺少有效的描述而成为英语统计机器翻译的一大挑战。针对长距离调序的可能途径:预调序,提出了一种基于神经网络的英文机辅翻译预调序模型。该模型在线性排序框架下结合神经网络建模,可以从大量样本数据中抽取句法和语义等有用信息,以预测不同语言的语序差异。最后在中文到英文的翻译机器和英文到中文的翻译机器上对该模型进行了实验。实验结果表明,该模型提高了系统性能,具有有效性。

Long-distance sequencing has become a major challenge for English statistical machine translation due to the lack of effective description.A preorder model of English machine-aided translation based on neural network is proposed in this paper to solve the possible path preconditioning for solving long-distance order.The model is based on neural network modeling in the linear sorting framework,and can extract useful information such as syntax and semantics from a large number of sample data to predict the difference of word order in different languages.The experiment was carried out on Chinese-English translation machine and English-Chinese translation machine.The experimental results show that the model improves the system performance and is effective.