Acta Scientiarum Naturalium Universitatis Pekinensis ›› 2023, Vol. 59 ›› Issue (1): 76-82.DOI: 10.13209/j.0479-8023.2022.066

Previous Articles     Next Articles

Exploration of Knowledge Driven Event Hyperbolic Embedding Temporal Relation Extraction Method

DUAN Jianyong1,2,†, DAI Shiwei1, WANG Hao1,2, HE Li1,2, LI Xin1,2   

  1. 1. School of Information Science and Technology, North China University of Technology, Beijing 100144 2. CNONIX National Standard Application and Promotion Lab, Beijing 100144
  • Received:2022-05-13 Revised:2022-07-25 Online:2023-01-20 Published:2023-01-20
  • Contact: DUAN Jianyong, E-mail: duanjy(at)


段建勇1,2,†, 戴诗伟1, 王昊1,2, 何丽1,2, 李欣1,2   

  1. 1. 北方工业大学信息学院, 北京 100144 2. CNONIX 国家标准应用与推广实验室, 北京 100144
  • 通讯作者: 段建勇, E-mail: duanjy(at)
  • 基金资助:


Aiming at the problem of asymmetric temporal relations of events, the event representation is mapped to hyperbolic space to extract temporal relations of events. The word embedding representation of the event is constructed by using the pre-trained word vector and external knowledge through simple operation. Experimental results on publicly released datasets show that the F1 value of the model is generally 2% higher than that of the baseline model, which can improve the effect of event temporal relation extraction.

Key words: event temporal sequence, relation extraction, hyperbolic space word embedding


针对事件时间关系不对称的问题, 采用将事件表示映射到双曲空间的方法, 进行事件时序关系抽取。通过简单的运算, 用预训练的词向量与外部知识构建事件的词嵌入表示。在公开发布的数据集上的实验结果表明, 模型的F1值比基线模型普遍高2%, 能够提升事件时序关系抽取的效果。

关键词: 事件时序, 关系抽取, 双曲空间词嵌入