一种基于IDLSTM+CRF的中文主地域抽取方法(A Chinese Main Location Extraction Method based on IDLSTM+CRF)
Yiqi Tong (童逸琦), Peigen Ye (叶培根), Biao Fu (付彪), Yidong Chen (陈毅东), Xiaodong Shi (史晓东)
Abstract
新闻文本通常会涉及多个地域,主地域则描述了文本舆情内容的地域属性,是进行舆情分析的关键属性。目前深度学习领域针对主地域自动抽取的研究还比较少。基于此,本文构建了一个基于IDLSTM+CRF的主地域抽取系统。该系统通过地名识别、主地域抽取、主地域补全三大模块实现对主地域标签的自动抽取和补全。在公开数据集上的实验结果表明,我们的方法在地名识别任务上要优于BiLSTM+CRF等模型。而对于主地域抽取任务,目前还没有标准的中文主地域评测集合。针对该问题,我们标注并开源了1226条验证集和1500条测试集。最终,我们的主地域抽取系统在两个集合上分别取得了91.7%和84.8%的抽取准确率,并成功运用于线上生产环境。- Anthology ID:
- 2021.ccl-1.71
- Volume:
- Proceedings of the 20th Chinese National Conference on Computational Linguistics
- Month:
- August
- Year:
- 2021
- Address:
- Huhhot, China
- Venue:
- CCL
- SIG:
- Publisher:
- Chinese Information Processing Society of China
- Note:
- Pages:
- 792–802
- Language:
- Chinese
- URL:
- https://aclanthology.org/2021.ccl-1.71
- DOI:
- Cite (ACL):
- Yiqi Tong, Peigen Ye, Biao Fu, Yidong Chen, and Xiaodong Shi. 2021. 一种基于IDLSTM+CRF的中文主地域抽取方法(A Chinese Main Location Extraction Method based on IDLSTM+CRF). In Proceedings of the 20th Chinese National Conference on Computational Linguistics, pages 792–802, Huhhot, China. Chinese Information Processing Society of China.
- Cite (Informal):
- 一种基于IDLSTM+CRF的中文主地域抽取方法(A Chinese Main Location Extraction Method based on IDLSTM+CRF) (Tong et al., CCL 2021)
- PDF:
- https://preview.aclanthology.org/paclic-22-ingestion/2021.ccl-1.71.pdf
- Data
- SQuAD