融合多元特征表示的藏文命名实体识别方法赵小兵∗2(Research on Tibetan Named Entity Recognition Using Multi-Feature Fusion Representation)

Ejian Cairang (俄见才让), Zhou Maoke (周毛克), Chen Bo (陈波), Zhao Xiaobing (赵小兵)


Abstract
“本文针对基于音节嵌入方式的藏文命名实体识别(TNER)中词汇信息和音节部件信息忽略的问题,提出了基于交叉Transformer架构的MECT-TL模型,融合了藏文音节信息、词汇信息和音节部件信息的多元数据特征。MECT-TL通过平面网络结构将藏文音节与词汇信息结合,并整合音节部件信息,有效提升了藏文实体识别的准确性。实验结果显示,相较于主流的TNER基准模型BiLSTM-CRF,本文模型在F1值上提高了5.14个百分点,与基于Transformer架构的TENER模型相比提高了4.18个百分点。这表明,融合藏文词汇和音节部件信息的方法可以显著提高TNER任务的性能。”
Anthology ID:
2024.ccl-1.26
Volume:
Proceedings of the 23rd Chinese National Conference on Computational Linguistics (Volume 1: Main Conference)
Month:
July
Year:
2024
Address:
Taiyuan, China
Editors:
Maosong Sun, Jiye Liang, Xianpei Han, Zhiyuan Liu, Yulan He
Venue:
CCL
SIG:
Publisher:
Chinese Information Processing Society of China
Note:
Pages:
340–351
Language:
Chinese
URL:
https://aclanthology.org/2024.ccl-1.26/
DOI:
Bibkey:
Cite (ACL):
Ejian Cairang, Zhou Maoke, Chen Bo, and Zhao Xiaobing. 2024. 融合多元特征表示的藏文命名实体识别方法赵小兵∗2(Research on Tibetan Named Entity Recognition Using Multi-Feature Fusion Representation). In Proceedings of the 23rd Chinese National Conference on Computational Linguistics (Volume 1: Main Conference), pages 340–351, Taiyuan, China. Chinese Information Processing Society of China.
Cite (Informal):
融合多元特征表示的藏文命名实体识别方法赵小兵∗2(Research on Tibetan Named Entity Recognition Using Multi-Feature Fusion Representation) (Cairang et al., CCL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.ccl-1.26.pdf