Computer Engineering and Applications ›› 2023, Vol. 59 ›› Issue (13): 49-60.DOI: 10.3778/j.issn.1002-8331.2208-0035

• Research Hotspots and Reviews • Previous Articles     Next Articles

Review of Research on Cross-Lingual Summarization

ZHENG Bofei, YUN Jing, LIU Limin, JIAO Lei, YUAN Jingshu   

  1. 1.College of Data Science and Application, Inner Mongolia University of Technology, Hohhot 010000, China
    2.Inner Mongolia Autonomous Region Engineering & Technology Research Center of Big Data Based Software Service, Hohhot 010000, China
  • Online:2023-07-01 Published:2023-07-01

跨语言摘要方法研究综述

郑博飞,云静,刘利民,焦磊,袁静姝   

  1. 1.内蒙古工业大学 数据科学与应用学院,呼和浩特 010000
    2.内蒙古自治区基于大数据的软件服务工程技术研究中心,呼和浩特 010000

Abstract: With the development of the Internet, articles in various languages spring up, in order for users to quickly understand the main content of the article, it is necessary to obtain the main text information between different languages. Cross-lingual summarization is a method that can reflect the main idea of text extracted from multiple languages by computer, which can effectively solve the above problems. From the perspective of the development of the cross-lingual summarization method, this paper firstly makes a comprehensive survey on the research work of the cross-lingual summarization method, and combs the development process of the cross-lingual summarization method. Secondly, the key technologies are discussed and analyzed, the differences and shortcomings of these methods are summarized, and the main line of research from the initial pipeline-based method to the end-to-end method emerged after the popularization of deep learning is outlined. Finally, the challenges and future research trends of cross-lingual summarization are analyzed and prospected.

Key words: cross-lingual summarization, deep learning, pipeline, end-to-end method, natural language processing

摘要: 随着互联网的发展,各种语言文章涌现。为了用户能快速地了解文章的主要内容,需要获取不同语言之间的文本主旨信息。跨语言摘要是利用计算机从多种语言文本中提炼出一种能反映文本主旨的方法。从跨语言摘要方法发展的角度切入,对跨语言摘要的研究工作进行了全面的调查,梳理了跨语言摘要方法的发展过程,定义了跨语言摘要的任务。对其关键技术进行讨论与分析,总结出这些方法的差异与不足,勾画出从最初的基于管道式方法到深度学习普及后出现的基于端到端的方法的研究主线,并对跨语言摘要数据集进行了分析总结。最后对跨语言摘要面临的挑战和未来研究趋势进行分析与展望。

关键词: 跨语言摘要, 深度学习, 管道式, 端到端方法, 自然语言处理