计算机工程与应用 ›› 2022, Vol. 58 ›› Issue (18): 43-58.DOI: 10.3778/j.issn.1002-8331.2203-0453

• 热点与综述 • 上一篇    下一篇

限定域中文事件抽取研究综述

李华昱,毕经纶,闫阳   

  1. 中国石油大学(华东) 计算机科学与技术学院,山东 青岛 266580
  • 出版日期:2022-09-15 发布日期:2022-09-15

Survey of Chinese Event Extraction in Restricted Domain

LI Huayu, BI Jinglun, YAN Yang   

  1. College of Computer Science and Technology, China University of Petroleum(East China), Qingdao, Shandong 266580, China
  • Online:2022-09-15 Published:2022-09-15

摘要: 事件抽取是信息抽取领域最具有挑战性的任务之一,也是知识图谱构建中的关键技术。事件抽取在阅读理解、文本摘要、问答系统等领域得到了广泛的应用。限定域事件抽取指的是系统所抽取的事件类型是预定义的,因此针对某一特定领域,限定域事件抽取的研究更具有研究价值,而且中文事件抽取由于中文语言特性问题,面临着较大挑战。介绍了中文事件抽取中面对的挑战,对限定域中文事件抽取的主要方法进行归纳总结,重点介绍了基于深度学习的方法,并总结了少样本情况下的事件抽取方法,介绍了中文事件抽取常用的数据集,展望了中文事件抽取未来的发展趋势。

关键词: 限定域, 中文事件抽取, 信息抽取

Abstract: Event extraction is one of the most challenging tasks in the field of information extraction, and it is also a key technology in the construction of knowledge map. Event extraction has been widely used in reading comprehension, text summary, question and answer system and other fields. Restricted domain event extraction refers to that the event types extracted by the system are predefined. Therefore, for a specific field, the research of restricted domain event extraction has more research value, and Chinese event extraction is facing great challenges due to the characteristics of Chinese language. This paper introduces the challenges faced in Chinese event extraction, and then summarizes the main methods of restricted domain Chinese event extraction, focusing on the methods based on deep learning, and summarizes the event extraction methods in the case of few samples, then introduces the commonly used data sets of Chinese event extraction, and finally looks forward to the future development trend of Chinese event extraction.

Key words: restricted domain, Chinese event extraction, information extraction