基于阅读理解框架的中文事件论元抽取(Chinese Event Argument Extraction using Reading Comprehension Framework)
Min Chen (陈敏), Fan Wu (吴凡), Zhongqing Wang (王中卿), Peifeng Li (李培峰), Qiaoming Zhu (朱巧明)
Abstract
传统的事件论元抽取方法把该任务当作句子中实体提及的多分类或序列标注任务,论元角色的类别在这些方法中只能作为向量表示,而忽略了论元角色的先验信息。实际上,论元角色的语义和论元本身有很大关系。对此,本文提议将其当作机器阅读理解任务,把论元角色表述为自然语言描述的问题,通过在上下文中回答这些问题来抽取论元。该方法更好地利用了论元角色类别的先验信息,在ACE2005中文语料上的实验证明了该方法的有效性。- Anthology ID:
- 2020.ccl-1.35
- Volume:
- Proceedings of the 19th Chinese National Conference on Computational Linguistics
- Month:
- October
- Year:
- 2020
- Address:
- Haikou, China
- Editors:
- Maosong Sun (孙茂松), Sujian Li (李素建), Yue Zhang (张岳), Yang Liu (刘洋)
- Venue:
- CCL
- SIG:
- Publisher:
- Chinese Information Processing Society of China
- Note:
- Pages:
- 376–389
- Language:
- Chinese
- URL:
- https://aclanthology.org/2020.ccl-1.35
- DOI:
- Cite (ACL):
- Min Chen, Fan Wu, Zhongqing Wang, Peifeng Li, and Qiaoming Zhu. 2020. 基于阅读理解框架的中文事件论元抽取(Chinese Event Argument Extraction using Reading Comprehension Framework). In Proceedings of the 19th Chinese National Conference on Computational Linguistics, pages 376–389, Haikou, China. Chinese Information Processing Society of China.
- Cite (Informal):
- 基于阅读理解框架的中文事件论元抽取(Chinese Event Argument Extraction using Reading Comprehension Framework) (Chen et al., CCL 2020)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2020.ccl-1.35.pdf