计算机工程与应用 ›› 2020, Vol. 56 ›› Issue (22): 25-32.DOI: 10.3778/j.issn.1002-8331.2001-0263

• 热点与综述 • 上一篇    下一篇

效率为先:机器翻译译后编辑技术综述

孟福永,唐旭日   

  1. 华中科技大学 外国语学院,武汉 430074
  • 出版日期:2020-11-15 发布日期:2020-11-13

Efficiency First:Reviewing Technologies of Machine Translation Post-Editing

MENG Fuyong, TANG Xuri   

  1. School of Foreign Languages, Huazhong University of Science and Technology, Wuhan 430074, China
  • Online:2020-11-15 Published:2020-11-13

摘要:

译后编辑是近年来随机器翻译发展新兴的计算机应用研究领域。综述了译后编辑所涉及的三方面的技术问题:自动译后编辑,自动机器翻译质量评估以及译后编辑工作量测量。自动译后编辑与自动机器翻译质量评估均应用自然语言处理领域相关技术,通过自动完成部分编辑工作、筛选机器翻译译文来降低译员工作量,从而提高效率。对译后编辑工作量的研究侧重于分析机器翻译译文的错误对译员工作量的影响,为译后编辑系统环境的设计提供理论依据。总体而言,译后编辑技术研究的主要目的是提高译员的工作效率。

关键词: 机器翻译(MT), 自动译后编辑(APE), 质量评估, 译后编辑工作量

Abstract:

Post-editing is a newly-rising research field with the development of Machine Translation(MT) in recent years. This paper reviews the technologies and studies concerning MT post-editing, including Automatic Post-Editing (APE), MT quality estimation and post-editing effort. APE and MT quality estimation utilize the technologies in natural language processing to rescue translators from repetitive work, thus improving the efficiency. Studies concerning post-editing effort focus on analyzing the impact of MT errors on post-editing effort, preparing theoretical foundations for designing post-editing environments. Overall, the primary aim of MT post-editing is to promote the efficiency of translators.

Key words: Machine Translation(MT), Automatic Post-Editing(APE), quality estimation, post-editing effort