[1]刘俊鹏,宋鼎新,张一鸣,等.多种数据泛化策略融合的神经机器翻译系统[J].江西师范大学学报(自然科学版),2020,(01):39-45.[doi:10.16357/j.cnki.issn1000-5862.2020.01.07]
 LIU Junpeng,SONG Dingxin,ZHANG Yiming,et al.The Neural Machine Translation System of Multiple Data Generalization Fusion[J].Journal of Jiangxi Normal University:Natural Science Edition,2020,(01):39-45.[doi:10.16357/j.cnki.issn1000-5862.2020.01.07]
点击复制

多种数据泛化策略融合的神经机器翻译系统()
分享到:

《江西师范大学学报》(自然科学版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2020年01期
页码:
39-45
栏目:
机器翻译
出版日期:
2020-02-10

文章信息/Info

Title:
The Neural Machine Translation System of Multiple Data Generalization Fusion
文章编号:
1000-5862(2020)01-0039-07
作者:
刘俊鹏宋鼎新张一鸣黄德根*
大连理工大学计算机科学与技术学院,辽宁 大连 116024
Author(s):
LIU JunpengSONG DingxinZHANG YimingHUANG Degen*
College of Computer Science and Technology,Dalian University of Technology,Dalian Liaoning 116024,China
关键词:
神经机器翻译 自注意力机制 数据泛化 中英翻译
Keywords:
neural machine translation self-attention data generalization Chinese-to-English translation
分类号:
TP 391
DOI:
10.16357/j.cnki.issn1000-5862.2020.01.07
文献标志码:
A
摘要:
在Transformer模型的基础上,该文从数据泛化、多样化解码策略和后处理方法3个方面进行改进.多种数据泛化策略融合方法对不同种类的稀疏词语进行识别、泛化和翻译,减少错译现象.利用检查点平均和模型集成等多样化解码策略进一步提升翻译效果.在CCMT 2019中英新闻领域翻译任务上的实验结果显示,改进后的方法在基线系统上的BLEU-SBP值提升了约1.85%.
Abstract:
The improvements on the Transformer baseline system are described from three aspects,including data generalization,multiple decoding strategies and post-processing.Multiple data generalization fusion method is used to recognize,generalize and translate different types of rare words,which reduces mistranslation in the neural machine translation.Multiple decoding strategies such as checkpoint averaging and model ensemble can further boost the translation performance.Experimental results on CCMT 2019 Chinese-English news translation task show that the proposed methods significantly improve translation performance by about 1.85% BLEU-SBP points than baseline system.

参考文献/References:

[1] Sutskever I,Vinyals O,Le Quoc V.Sequence to sequence learning with neural networks[EB/OL].[2019-06-17].https://arxiv.org/abs/1409.3215.
[2] Hochreiter S,Schmidhuber J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[3] Cho K,Merrienboer B,Gulcehre C,et al.Learning phrase representations using RNN encoder-decoder for statistical machine translation[EB/OL].[2019-06-17].https://arxiv.org/abs/1406.1078.
[4] Bahdanau D,Cho K,Bengio Y.Neural machine translation by jointly learning to align and translate[EB/OL].[2019-06-17].https://arxiv.org/abs/1409.0473v2.
[5] Koehn P,Och F J,Marcu D.Statistical phrase-based translation[EB/OL].[2019-06-17].http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.69.4271.
[6] Chiang D.A hierarchical phrase-based model for statistical machine translation[EB/OL].[2019-06-17].https://dl.acm.org/citation.cfm?doid=1219840.1219873.
[7] Vaswani A,Shazeer N,Parmar N,et al.Attention is all you need[EB/OL].[2019-06-17].https://arxiv.org/abs/1706.03762.
[8] Ba J L,Kiros J R,Hinton G E.Layer normalization[EB/OL].[2019-06-17].http://arxiv.org/abs/1607.06450.
[9] He Kaiming,Zhang Xiangyu,Ren Shaoqing,et al.Identity mappings in deep residual networks[C].Amsterdam:Springer,2016:630-645.
[10] Xiao Tong,Zhu Jingbo,Zhang Hao,et al.NiuTrans:an open source toolkit for phrase-based and syntax-based machine translation[C].Jeju:Association for Computational Linguistics,2012:19-24.
[11] Senntich R,Haddow B,Brich A.Neural machine translation of rare words with subword units[C].Berlin:Association for Computational Linguistics,2016:1715-1725.
[12] Huang Fei,Vogel S,Waibel A.Automatic extraction of named entity translingual equivalence based on multi-feature cost minimization[C].Sapporo:Association for Computational Linguistics,2003:9-16.
[13] Feng Donghui,Lü Yajuan,Zhou Ming.A new approach for English-Chinese named entity alignment[C].Barcelona:Association for Computational Linguistics,2004:372-379.
[14] Sennrich R,Haddow B,Birch A.Edinburgh neural machine translation systems for WMT 16[C].Berlin:Association for Computational Linguistics,2016:371-376.
[15] Sennrich R,Birch A,Currey A,et al.The University of Edinburgh's Neural MT Systems for WMT17[C].Copenhagen:Association for Computational Linguistics,2017:389-399.
[16] Wang Yuguang,Cheng Shanbo,Jiang Liyang,et al.Sogou neural machine translation systems for wmt17[C].Copenhagen:Association for Computational Linguistics,2017:410-415.
[17] Kingma D P,Ba J.Adam:a method for stochastic optimization[EB/OL].[2019-06-17].https://arxiv.org/abs/1412.6980.
[18] Wu Yonghui,Schuster M,CHEN Zhifeng,et al.Google's neural machine translation system:bridging the gap between human and machine translation[EB/OL].[2016-10-08][2019-08-29].http://arxiv.org/pdf/1609.08144.pdf.
[19] Chiang D,Deneefe S,Chan Yee Seng,et al.Decomposability of translation metrics for improved evaluation and efficient algorithms[C].Honolulu:Association for Computational Linguistics,2008:610-619.

相似文献/References:

[1]王 坤,殷明明,俞鸿飞,等.低资源维汉神经机器翻译研究[J].江西师范大学学报(自然科学版),2019,(06):638.[doi:10.16357/j.cnki.issn1000-5862.2019.06.13]
 WANG Kun,YIN Mingming,YU Hongfei,et al.The Study on Low-Resource Uygur-Chinese Neural Machine Translation[J].Journal of Jiangxi Normal University:Natural Science Edition,2019,(01):638.[doi:10.16357/j.cnki.issn1000-5862.2019.06.13]
[2]谷舒豪,单 勇,谢婉莹,等.基于数据增强及领域适应的神经机器翻译技术[J].江西师范大学学报(自然科学版),2019,(06):643.[doi:10.16357/j.cnki.issn1000-5862.2019.06.14]
 GU Shuhao,SHAN Yong,XIE Wanying,et al.The Neural Machine Translation Based on Data Augmentation and Domain Adaptation Technology[J].Journal of Jiangxi Normal University:Natural Science Edition,2019,(01):643.[doi:10.16357/j.cnki.issn1000-5862.2019.06.14]
[3]钟茂生,吴佳华,罗 玮,等.面向低资源命名实体识别的BiLSTM-Att-BCRF模型[J].江西师范大学学报(自然科学版),2022,(05):460.[doi:10.16357/j.cnki.issn1000-5862.2022.05.04]
 ZHONG Maosheng,WU Jiahua,LUO Wei,et al.The BiLSTM-Att-BCRF Model for Low Resource Named Entity Recognition[J].Journal of Jiangxi Normal University:Natural Science Edition,2022,(01):460.[doi:10.16357/j.cnki.issn1000-5862.2022.05.04]
[4]赵 阳,周 龙,王 迁,等.民汉稀缺资源神经机器翻译技术研究[J].江西师范大学学报(自然科学版),2019,(06):630.[doi:10.16357/j.cnki.issn1000-5862.2019.06.12]
 ZHAO Yang,ZHOU Long,WANG Qian,et al.The Study on Ethnic-to-Chinese Scare-Resource Neural Machine Translation[J].Journal of Jiangxi Normal University:Natural Science Edition,2019,(01):630.[doi:10.16357/j.cnki.issn1000-5862.2019.06.12]

备注/Memo

备注/Memo:
收稿日期:2019-09-08
基金项目:国家自然科学基金(61672127)资助项目.
通信作者:黄德根(1965-),男,福建邵武人,教授,博士生导师,主要从事自然语言处理、神经机器翻译方面的研究.E-mail:huangdg@dlut.edu.cn
更新日期/Last Update: 2020-02-10