Knowledge-Enriched Event Causality Identification via Latent Structure Induction Networks
文献类型:会议论文
作者 | Pengfei Cao1,2; Xinyu Zuo1,2; Yubo Chen1,2; Kang Liu1,2; Jun Zhao1,2; Yuguang Chen3; Weihua Peng3 |
出版日期 | 2021 |
会议日期 | August 1-6, 2021 |
会议地点 | Bangkok, Thailand (Online) |
英文摘要 | Identifying causal relations of events is an important task in natural language processing area. However, the task is very challenging, because event causality is usually expressed in diverse forms that often lack explicit causal clues. Existing methods cannot handle well the problem, especially in the condition of lacking training data. Nonetheless, humans can make a correct judgement based on their background knowledge, including descriptive knowledge and relational knowledge. Inspired by it, we propose a novel Latent Structure Induction Network (LSIN) to incorporate the external structural knowledge into this task. Specifically, to make use of the descriptive knowledge, we devise a Descriptive Graph Induction module to obtain and encode the graph-structured descriptive knowledge. To leverage the relational knowledge, we propose a Relational Graph Induction module which is able to automatically learn a reasoning structure for event causality reasoning. Experimental results on two widely used datasets indicate that our approach significantly outperforms previous state-of-the-art methods. |
语种 | 英语 |
源URL | [http://ir.ia.ac.cn/handle/173211/44857] |
专题 | 模式识别国家重点实验室_自然语言处理 |
作者单位 | 1.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 2.School of Artificial Intelligence, University of Chinese Academy of Sciences 3.Beijing Baidu Netcom Science Technology Co., Ltd |
推荐引用方式 GB/T 7714 | Pengfei Cao,Xinyu Zuo,Yubo Chen,et al. Knowledge-Enriched Event Causality Identification via Latent Structure Induction Networks[C]. 见:. Bangkok, Thailand (Online). August 1-6, 2021. |
入库方式: OAI收割
来源:自动化研究所
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。