基于现代智能识别技术的英语机器翻译模型English machine translation model based on modern intelligent recognition technology
张帆
摘要(Abstract)：
基于句法分析的英语机器翻译方法无法解决智能识别技术中海量英语语言的部分结构歧义导致机器翻译准确度低的问题。因此,在分析智能机器翻译工具辅助英语翻译应用的基础上,设计与实现基于现代智能识别技术的英语机器翻译模型。依据英文句子产生中文句子和对齐过程,基于英文句子获取中文句子长度、首个中文词串的链接位置,获取总体中文句子以及机器翻译对在句子中的次数。采用基于最大熵的统计机器翻译方法,通过直接最大熵模型训练得到相关参数,获取不同英语语言特征间的最佳组合方式,解决海量英语语言中的部分结构歧义问题,提高英语机器翻译的准确度。实验结果表明,所设计的英语机器翻译模型,具有较高的翻译准确度和稳定性。
关键词(KeyWords)： 智能识别技术;英语翻译;机器翻译模型;结构歧义;最大熵;翻译准确度
基金项目(Foundation): 四川省教育厅自筹科研项目(15ZB0276)~~
作者(Author): 张帆
DOI: 10.16652/j.issn.1004-373x.2018.16.038
参考文献(References)：
- [1]赵静.基于统计的汉英机器翻译技术的研究[J].电子设计工程,2016,24(21):69-71.ZHAO Jing.Research on the technology of C-E machine translation based on statistics[J].Electronic design engineering,2016,24(21):69-71.
- [2]宋柔,葛诗利.面向篇章机器翻译的英汉翻译单位和翻译模型研究[J].中文信息学报,2015,29(5):125-135.SONG Rou,GE Shili.English-Chinese translation unit and translation model for discourse-based machine translation[J].Journal of Chinese information processing,2015,29(5):125-135.
- [3]唐海庆,熊德意.基于选择偏向性的统计机器翻译模型[J].北京大学学报(自然科学版),2016,52(1):127-133.TANG Haiqing,XIONG Deyi.A selectional preference based translation model for SMT[J].Acta Scientiarum Naturalium Universitatis Pekinensis,2016,52(1):127-133.
- [4]杨呈永.物联网虚假数据特征自适应检测方法仿真[J].计算机仿真,2017,34(4):276-279.YANG Chengyong.Simulation of false data feature adaptive detection in Internet of Things[J].Computer simulation,2017,34(4):276-279.
- [5]HERMANTO A,ADJI T B,SETIAWAN N A.Recurrent neural network language model for English-Indonesian machine translation:experimental study[C]//Proceedings of International Conference on Science in Information Technology.Yogyakarta:IEEE,2016:132-136.
- [6]ESCOLANO C,COSTA-JUSSàM R,FONOLLOSA J A R.The TALP-UPC neural machine translation system for German/Finnish-English using the inverse direction model in rescoring[C]//Proceedings of 2nd Conference on Machine Translation.Copenhagen:Association for Computational Linguistics,2017:283-287.
- [7]BASTAN M,KHADIVI S,HOMAYOUNPOUR M M.Neural machine translation on scarce-resource condition:a case-study on Persian-English[C]//Proceedings of Iranian Conference on Electrical Engineering.Tehran:IEEE,2017:1485-1490.
- [8]胡滨.基于信息共享空间的科技博物馆服务模式创新[J].科技通报,2015,31(11):277-282.HU Bin.Innovation in service model of science and technology museum based on information commons[J].Bulletin of science and technology,2015,31(11):277-282.
- [9]何向真,万福成,于洪志,等.基于藏语语义分析的机器翻译技术研究[J].计算机工程与应用,2015,51(15):134-137.HE Xiangzhen,WAN Fucheng,YU Hongzhi,et al.Machine translation technology based on Tibetan semantic parsing[J].Computer engineering and applications,2015,51(15):134-137.
- [10]高鹏,黄瑛.矿物加工工程英语的特征及翻译[J].中国科技翻译,2016,29(2):15-18.GAO Peng,HUANG Ying.The characteristics and translation of mineral processing engineering English[J].Chinese science&technology translators journal,2016,29(2):15-18.