基于相似度模型的英语机器翻译研究  

Research on English machine translation based on similarity model

作　　者：巫奕君[1] 秦永红[2] Wu Yijun;Qin Yonghong(City College,Xi'an Jiaotong University,Xi'an,Shaanxi 710018;School of Electrical Engineering,Southwest Jiaotong University,Chengdu,Sichuan 610031)

机构地区：[1]西安交通大学城市学院,陕西西安710018 [2]西南交通大学电气工程学院,四川成都610031

出　　处：《现代科学仪器》2020年第6期159-162,共4页Modern Scientific Instruments

摘　　要：本文基于相似度模型,对传统英语机器翻译存在的问题进行了研究。论文在分析语义相似度概念的基础之上提出了语义合成算法,并采用神经网络不断地训练得到最佳的语义向量权重系数,同时在原有英语机器翻译模型的基础之上将英语翻译的评价结果反馈给语言学专家,不断地更新语料库,对原有的语料库及时更新和丰富。通过英语翻译试验,将改进的英语机器翻译模型和层次短语翻译模型、单词分布语义信息模型进行对比。试验结果表明改进的英语机器翻译模型对英语翻译的准确度更高,翻译性能得到明显改善。本论文的研究对提升英语翻译准确度具有一定的参考价值。Based on the similarity model,this paper studies the problems of traditional English machine translation.Based on the analysis of the concept of semantic similarity,this paper proposes a semantic synthesis algorithm,and uses neural network to train the best weight coefficient of semantic vector.Meanwhile,based on the original English machine translation model,the evaluation results of English translation are fed back to the experts of linguistics to update and enrich the original corpus in time.Through English translation experiments,the improved English machine translation model is compared with hierarchical phrase translation model and word distribution semantic information model.The experimental results show that the improved English machine translation model has higher accuracy and better translation performance.The study of this paper has a certain reference value to improve the accuracy of English translation.

关 键 词：机器翻译 语义相识度 语义合成算法 

分 类 号：TP391.2[自动化与计算机技术—计算机应用技术]