维吾尔语复杂形态对汉维机器翻译的影响研究

doi:10.19678/j.issn.1000-3428.0053080

计算机工程 ›› 2020, Vol. 46 ›› Issue (2): 309-314. doi: 10.19678/j.issn.1000-3428.0053080

维吾尔语复杂形态对汉维机器翻译的影响研究

穆妮热·穆合塔尔^1,2,3, 李晓^1,2, 杨雅婷^1,2

1. 中国科学院新疆理化技术研究所, 乌鲁木齐 830011;
2. 中国科学院大学, 北京 100049;
3. 新疆民族语音语言信息处理实验室, 乌鲁木齐 830011

收稿日期:2018-11-07 修回日期:2019-03-04 发布日期:2019-03-14
作者简介:穆妮热·穆合塔尔(1989-),女,博士研究生,主研方向为自然语言处理、机器翻译;李晓(通信作者),研究员、博士生导师;杨雅婷,副研究员、博士。
基金资助:
国家自然科学基金（U1703133）；中科院西部之光人才培养引进计划（2017-XBQNXZ-A-005）；中国科学院青年创新促进会项目（2017472）；新疆维吾尔自治区重大科技专项（2016A03007-3）；新疆维吾尔自治区高层次人才引进工程（Y839031201）。

Research on Influence of Uyghur Complex Morphology on Chinese-Uyghur Machine Translation

MUNIRE·Muhetare^1,2,3, LI Xiao^1,2, YANG Yating^1,2

1. Xinjiang Technical Institute of Physics and Chemistry, Chinese Academy of Sciences, Urumqi 830011, China;
2. University of Chinese Academy of Sciences, Beijing 100049, China;
3. Xinjiang Laboratory of Minority Speech and Language Information Processing, Urumqi 830011, China

Received:2018-11-07 Revised:2019-03-04 Published:2019-03-14

摘要/Abstract

摘要： 维吾尔语形态较为复杂，构形词缀在维吾尔语中占有重要地位，其语法与汉语有较大差别。针对维吾尔语的形态特点，分析汉语端到维吾尔语端在统计机器翻译中维吾尔语词缀的作用，搭建基于短语的汉维统计机器翻译系统，对词级粒度、词干级粒度、最大词干级粒度、词干-词缀级粒度、词干-词尾级粒度的汉维平行语料库进行对比实验，研究不同粒度的维吾尔语对汉维机器翻译中的词语对齐质量和语言模型质量的影响。实验结果表明，在上述5种粒度的维吾尔语语料中，基于词干的维吾尔语和基于词干-词尾的维吾尔语目标端语料的翻译质量明显提高。

关键词: 维吾尔语形态, 构形词缀, 词缀粒度, 统计机器翻译, 翻译质量

Abstract: The Uyghur morphology is comparatively complex and the configuration affix plays a significant role in Uyghur,which is grammatically very different from Chinese.Aiming at the morphology characteristics of Uyghur,this paper analyzes the function of Uyghur affix in statistical machine translation from Chinese to Uyghur.A phrase-based Chinese-Uyghur statistical translation system is built to conduct comparative experiments on Chinese-Uyghur parallel corpus with different levels of granularity,such as the word level granularity,the stem level granularity,the maximum stem level granularity,the stem-affix level granularity and the stem-suffix level granularity.Then the influence of Uyghur with different granularity on words alignment quality and language model quality in Chinese-Uyghur machine translation is studied.Experimental results show that the translation quality of the stem-based and the stem-suffix based Uyghur target corpus is significantly improved.

Key words: Uyghur morphology, configuration affix, affix granularity, statistical machine translation, translation quality

中图分类号:

TP391

穆妮热·穆合塔尔, 李晓, 杨雅婷. 维吾尔语复杂形态对汉维机器翻译的影响研究[J]. 计算机工程, 2020, 46(2): 309-314.

MUNIRE·Muhetare, LI Xiao, YANG Yating. Research on Influence of Uyghur Complex Morphology on Chinese-Uyghur Machine Translation[J]. Computer Engineering, 2020, 46(2): 309-314.

http://www.ecice06.com/CN/Y2020/V46/I2/309

图/表 12

20200218144641

20200218144644

20200218144648

20200218144651

20200218144654

20200218144658

20200218144702

20200218144705

20200218144709

20200218144713

20200218144717

20200218144720

参考文献

[1] WANG Zhixiang.Research on key technologies of machine translation for morphologically rich languages[D].Beijing:University of Chinese Academy of Sciences,2013.(in Chinese) 王志洋.面向形态丰富语言的机器翻译关键技术研究[D].北京:中国科学院大学,2013.
[2] LITIFU·Tohuti.From phrase structure to minimalism:syntactic structure in altai[M].Beijing:Minzu University of China Press,2004.(in Chinese) 力提甫·托乎提.从短语结构到最简方案:阿尔泰语言的句法结构[M].北京:中央民族大学出版社,2004.
[3] LITIFU·Tohuti.Study of generative syntax in Uyghur and other Altai languages[M].Beijing:Ethnic Publishing House,2001.(in Chinese) 力提甫·托乎提.维吾尔语及其他阿尔泰语言的生成句法研究[M].北京:民族出版社,2001.
[4] XUEHELAITI·Miliwan,LIU Kai,TURGUN·Ibrahim.Chinese-Uyghur machine translation based on smallest translation units of stems and suffixes[J].Journal of Chinese Information Processing,2015,29(3):201-206.(in Chinese) 米莉万·雪合来提,刘凯,吐尔根·依布拉音.基于维吾尔语词干词缀粒度的汉维机器翻译[J].中文信息学报,2015,29(3):201-206.
[5] LI Wen.Research on construction method of morphological asymmetric Chinese-Mongolian statistical machine translation model[J].Hefei:University of Science and Technology of China,2014.(in Chinese) 李文.形态非对称汉蒙统计机器翻译模型构造方法研究[D].合肥:中国科学技术大学,2014.
[6] TUERXUN Aizizi.Research on Uyghur morphological analysis for machine translation[D].Beijing:University of Chinese Academy Sciences,2017.(in Chinese) 艾孜孜·吐尔逊.面向机器翻译的维吾尔语形态分析研究[D].北京:中国科学院大学,2017.
[7] BAO Meirong.Chinese Mongolian statistical machine translation with morphological analysis[D].Hohhot:Inner Mongolia Normal University,2012.(in Chinese) 宝美荣.融入形态学分析的汉蒙统计机器翻译研究[D].呼和浩特:内蒙古师范大学,2012
[8] AZRAGUL,NUERAIHEMAITI,AIBAIDOULA·Yusufu.Research on key technology for statistics of modern Uyghur language[J].Journal of Chinese Information Processing,2014,28(5):192-197.(in Chinese) 艾孜尔古丽,努尔艾合买提,玉素甫·艾白都拉.现代维吾尔语常用词统计关键技术研究[J].中文信息学报,2014,28(5):192-197.
[9] CHEN Mengyun,GAO Jianhua.Static test cases sorting method based on cyclomatic complexity[J].Computer Applications and Software,2016,33(1):1-3,15.(in Chinese) 陈梦云,高建华.基于圈复杂度的静态测试用例排序方法[J].计算机应用与软件,2016,33(1):1-3,15.
[10] FENG Qiuxiang,WANG Rongpei.Database semantics and natural language communication[J].Computer Science,2011,38(11):187-190.(in Chinese) 冯秋香,汪榕培.数据库语义学与自然语言交流[J].计算机科学,2011,38(11):187-190.
[11] TIEMUER Haimiti.Modern Uyghur grammar[M].Beijing:Ethnic Publishing House,2011.(in Chinese) 哈米提·铁木尔.现代维吾尔语语法[M].北京:民族出版社,2011.
[12] WUMAIER·Aishan.Research on key technologies of lexical and syntactic analysis of Uyghur language[D].Urumqi:Xinjiang University,2010.(in Chinese) 艾山·吾买尔.维吾尔语词法句法分析关键技术的研究[D].乌鲁木齐:新疆大学,2010.
[13] ZHOU Ran,SONG Huazhu,LIU Xiang.Research on settlement particle recognition based on Haar-PP mixed feature extraction[J].Computer Engineering,2018,44(12):251-257.(in Chinese) 周冉,宋华珠,刘翔.基于Haar-PP混合特征提取的沉降颗粒识别研究[J].计算机工程,2018,44(12):251-257.
[14] DONG Xinghua,ZHOU Junlin,GUO Shusheng,et al.Phrase-based Chinese-Uyghur/Uyghur-Chinese statistical machine translation[J].Computer Engineering,2011,37(9):16-18.(in Chinese) 董兴华,周俊林,郭树盛,等.基于短语的汉维(维汉)统计机器翻译[J].计算机工程,2011,37(9):16-18.
[15] ARIANNA B,MARCELLO F.Morphological pre-processing for turkish to english statistical machine translation[C]//Proceedings of IWSLT’09.Tokyo,Japan:[s.n.],2009:243-252.
[16] OFLAZER K,DURGAR E K.Exploring different representational units in English to statistical machine translation[C]//Proceedings of the 2nd Workshop on Statistical Machine Translation.Prague,CzechRepublic:[s.n.],2007:332-345.
[17] TACHBELIE M Y,ABETA S T,Besacier L.Using different acoustic,lexical,and language modeling units for ASR of an under-resourced language-Amharic[J].Speech Communication,2014,56(1):181-194.
[18] NEUBIG G.Unsupervised learning of lexical infor-mation for language processing systems[D].Kyoto,Japan:Kyoto University,2012.
[19] ARISOY E,SARACLAR M,ROARK B,et al.Discriminative language modeling with linguistic and statistically derived features[J].IEEE Transactions on Audio,Speech and Language Processing,2012,20(2):540-550.
[20] DURRETT G,DENERO J.Supervised learning of complete morphological paradigms[C]//Proceedings of NAACL-HLT’13.Atlanta,USA:[s.n.],2013:1185-1195.

选择文件类型/文献管理软件名称

选择包含的内容

维吾尔语复杂形态对汉维机器翻译的影响研究

Research on Influence of Uyghur Complex Morphology on Chinese-Uyghur Machine Translation

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献

相关文章 5

编辑推荐

Metrics

本文评价

[1]	周珂,余正涛,高盛祥. 融合主题的汉越冶金领域统计机器翻译方法[J]. 计算机工程, 2017, 43(12): 179-183.
[2]	米莉万.雪合来提,麦热哈巴.艾力，吐尔根.依布拉音，姜文斌. 维吾尔语词尾对汉维统计机器翻译影响的研究[J]. 计算机工程, 2014, 40(3): 224-227.
[3]	陈丽娟, 张恒, 董兴华, 吐尔洪?吾司曼, 周俊林. 基于句法调序的汉维统计机器翻译[J]. 计算机工程, 2012, 38(3): 169-171,175.
[4]	董兴华, 陈丽娟, 周喜, 周俊林, 吐尔洪?吾司曼. 汉维统计机器翻译中的形态学处理[J]. 计算机工程, 2011, 37(12): 150-152.
[5]	陈伟;黄蕾;刘峰;赵志宏. 双语平行网页挖掘系统的设计与实现[J]. 计算机工程, 2009, 35(14): 267-269.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

维吾尔语复杂形态对汉维机器翻译的影响研究

Research on Influence of Uyghur Complex Morphology on Chinese-Uyghur Machine Translation

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献

相关文章 5

编辑推荐

Metrics

本文评价