吉林大学学报(理学版) ›› 2023, Vol. 61 ›› Issue (5): 1095-1102.

• • 上一篇    下一篇

基于关系过滤和实体对标注的中文关系抽取方法

刘旭, 杨航, 张啸成, 张永刚   

  1. 吉林大学 计算机科学与技术学院, 符号计算与知识工程教育部重点实验室, 长春 130012
  • 收稿日期:2022-11-10 出版日期:2023-09-26 发布日期:2023-09-26
  • 通讯作者: 张永刚 E-mail:zhangyg@jlu.edu.cn

Chinese Relation Extraction Method Based on Relation Filtering and Entity Pair Tagging

LIU Xu, YANG Hang, ZHANG Xiaocheng, ZHANG Yonggang   

  1. Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, College of Computer Science and Technology, Jilin University, Changchun 130012, China
  • Received:2022-11-10 Online:2023-09-26 Published:2023-09-26

摘要: 针对关系三元组抽取任务中的冗余关系问题和实体重叠问题, 提出一种基于关系过滤器的二维实体对标注方案(RF2DTagging). RF2DTagging模型由两部分组成: 1) 用于过滤冗余关系的关系过滤器(relation filter); 2) 能有效解决各种实体重叠问题的二维实体对标注方案(2D entity-pair tagging scheme). 为进一步验证RF2DTagging模型, 在3个公开的中文关系抽取数据集(CCKS2019-Task3,CMeIE和DuIE2.0)上进行实验. 实验结果表明, 该模型能有效解决上述两个问题, 且总体性能比对比模型更好.

关键词: 中文关系抽取, 知识图谱, 二维实体对标注, 自然语言处理

Abstract: Aiming at the redundant relations and entity overpalling problems in  the task of relational triple extraction,  we proposd a 2D entity pair tagging scheme based on the relation filter (RF2DTagging).  RF2DTagging model consisted of two parts: 1) A relation filter for filtering redundant relations, and 2) a 2D entity pair tagging scheme that could effectively solve various entity overlapping problems. To further validate the RF2DTagging model, we conducted experiments on three public Chinese relation extraction datasets CCKS2019-Task3, CMeIE and DuIE2.0. The experimental results show that the  model can effectively solve the above two problems,  and the overall performance is better than the comparison model.

Key words: Chinese relation extraction, knowledge graph, 2D entity pair tagging, natural language processing

中图分类号: 

  • TP181