Abstract
中文社交媒体命名实体识别由于其领域特殊性,一直广受关注。非正式且无结构的微博文本存在以下两个问题:一是词语边界模糊;二是语料规模有限。针对问题一,本文将同维度的字词进行融合,获得丰富的文本序列表征;针对问题二,提出了基于Star-Transformer框架的命名实体识别模型,借助星型拓扑结构更好地捕获动态特征;同时利用高速网络优化Star-Transformer中的信息桥接,提升模型的鲁棒性。本文提出的轻量级命名实体识别模型取得了目前Weibo语料上最好的效果。- Anthology ID:
- 2020.ccl-1.37
- 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:
- 402–413
- Language:
- Chinese
- URL:
- https://aclanthology.org/2020.ccl-1.37
- DOI:
- Cite (ACL):
- Chun Chen, Mingyang Li, and Fang Kong. 2020. 面向微博文本的融合字词信息的轻量级命名实体识别(Lightweight Named Entity Recognition for Weibo Based on Word and Character). In Proceedings of the 19th Chinese National Conference on Computational Linguistics, pages 402–413, Haikou, China. Chinese Information Processing Society of China.
- Cite (Informal):
- 面向微博文本的融合字词信息的轻量级命名实体识别(Lightweight Named Entity Recognition for Weibo Based on Word and Character) (Chen et al., CCL 2020)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2020.ccl-1.37.pdf