统计机器翻译领域自适应方法比较研究 |
| |
引用本文: | 丁亮,李颖,何彦青.统计机器翻译领域自适应方法比较研究[J].情报工程,2016,2(4):080-088. |
| |
作者姓名: | 丁亮 李颖 何彦青 |
| |
作者单位: | 中国科学技术信息研究所 北京 100038 |
| |
基金项目: | 国家自然科学基金项目:(61303152、 71503240 和 71403257),中国科学技术信息研究所重点工作项目: (ZD2016-05) |
| |
摘 要: | 统计机器翻译常常面临训练数据与待翻译文本领域不一的问题,从而影响了翻译的性能,因此领域自适应一直是研究者关注的课题。本文以传统自适应方法和现行的机器学习方法为框架,介绍了近年来统计机器翻译领域自适应研究的进展。分析了各类研究方法的优缺点并对未来研究做出展望。
|
关 键 词: | 统计机器翻译 领域自适应 语料选取 翻译性能改进 |
Comparison Study of Domain Adaption Methods in Statistical Machine Translation |
| |
Authors: | DING Liang LI Ying and HE YanQing |
| |
Institution: | Institute of Scientific and Technical Information of China,Institute of Scientific and Technical Information of China and Institute of Scientific and Technical Information of China |
| |
Abstract: | Statistical machine translation(SMT)is often faced with the problem of different domains between the training data set and test data set,which affecting the performance of translation,therefore domain adaptation has been a subject of concern.In this paper,we constructed the domain adaptation research framework-the traditional adaptive methods and the existing machine learning methods,and introduced the research progress in recent years.We analyzed the advantages and disadvantages of each method and made a prospect for the future research. |
| |
Keywords: | Statistical machine translation domain adaptation corpus selection translation performance improvement |
本文献已被 万方数据 等数据库收录! |
| 点击此处可从《情报工程》浏览原始摘要信息 |
| 点击此处可从《情报工程》下载免费的PDF全文 |