CASIA OpenIR  > 模式识别国家重点实验室  > 自然语言处理
Making Language Model as Small as Possible in Statistical Machine Translation
Yang Liu1; Jiajun Zhang1; Jie Hao2; Dakun Zhang2
2014
Conference Name第十届全国机器翻译研讨会(The 10th China Workshop on Machine Translation,CWMT2014)
Conference Date2014-11
Conference Place澳门
Abstract

As one of the key components, n-gram language model is most
frequently used in statistical machine translation. Typically, higher order of the language model leads to better translation performance. However, higher order of the n-gram language model requires much more monolingual training data to avoid data sparseness. Furthermore, the model size increases exponentially when the n-gram order becomes higher and higher. In this paper, we investigate the language model pruning techniques that aim at making the model size as small as possible while keeping the translation quality. According to our investigation, we further propose to replace the higher order n-grams with a low-order cluster-based language model. The extensive experiments show that
our method is very effective.
 

Indexed ByEI
Language英语
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/23996
Collection模式识别国家重点实验室_自然语言处理
Corresponding AuthorYang Liu
Affiliation1.NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China
2.Toshiba (China) R&D Center, Beijing, China
First Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Corresponding Author AffilicationChinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
Recommended Citation
GB/T 7714
Yang Liu,Jiajun Zhang,Jie Hao,et al. Making Language Model as Small as Possible in Statistical Machine Translation[C],2014.
Files in This Item: Download All
File Name/Size DocType Version Access License
1_CWMT2014.pdf(263KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Yang Liu]'s Articles
[Jiajun Zhang]'s Articles
[Jie Hao]'s Articles
Baidu academic
Similar articles in Baidu academic
[Yang Liu]'s Articles
[Jiajun Zhang]'s Articles
[Jie Hao]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Yang Liu]'s Articles
[Jiajun Zhang]'s Articles
[Jie Hao]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 1_CWMT2014.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.