基于BiLSTM-CRF的社会突发事件研判方法(Social Emergency Event Judgement based on BiLSTM-CRF)

Huijun Hu (胡慧君), Cong Wang (王聪), Jianhua Dai (代建华), Maofu Liu (刘茂福)


Abstract
社会突发事件的分类和等级研判作为应急处置中的一环,其重要性不言而喻。然而,目前研究多数采用人工或规则的方法识别证据进行研判,由于社会突发事件的构成的复杂性和语言描述的灵活性,这对于研判证据识别有很大局限性。本文参考“事件抽取”思想,事件类型和研判证据作为事件中元素,以BiLSTM-CRF方法细粒度的识别,并将二者结合,分类结果作为等级研判的输入,识别出研判证据。最终将识别结果结合注意力机制进行等级研判,通过对研判证据的精准识别从而来增强等级研判的准确性。实验表明,相比人工或规则识别研判证据,本文提出的方法有着更好的鲁棒性,社会突发事件研判时也达到了较好的效果。 关键词:事件分类 ;研判证据识别 ;等级研判 ;BiLSTM-CRF
Anthology ID:
2020.ccl-1.62
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:
667–675
Language:
Chinese
URL:
https://aclanthology.org/2020.ccl-1.62
DOI:
Bibkey:
Cite (ACL):
Huijun Hu, Cong Wang, Jianhua Dai, and Maofu Liu. 2020. 基于BiLSTM-CRF的社会突发事件研判方法(Social Emergency Event Judgement based on BiLSTM-CRF). In Proceedings of the 19th Chinese National Conference on Computational Linguistics, pages 667–675, Haikou, China. Chinese Information Processing Society of China.
Cite (Informal):
基于BiLSTM-CRF的社会突发事件研判方法(Social Emergency Event Judgement based on BiLSTM-CRF) (Hu et al., CCL 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.ccl-1.62.pdf