基于Self-Attention的句法感知汉语框架语义角色标注(Syntax-Aware Chinese Frame Semantic Role Labeling Based on Self-Attention)

Xiaohui Wang (王晓晖), Ru Li (李茹), Zhiqiang Wang (王智强), Qinghua Chai (柴清华), Xiaoqi Han (韩孝奇)


Abstract
框架语义角色标注(Frame Semantic Role Labeling, FSRL)是基于FrameNet标注体系的语义分析任务。语义角色标注通常对句法有很强的依赖性,目前的语义角色标注模型大多基于双向长短时记忆网络Bi-LSTM,虽然可以获取句子中的长距离依赖信息,但无法很好获取句子中的句法信息。因此,引入self-attention机制来捕获句子中每个词的句法信息。实验结果表明,该模型在CFN(Chinese FrameNet,汉语框架网)数据集上的F1达到83.77%,提升了近11%。
Anthology ID:
2020.ccl-1.57
Volume:
Proceedings of the 19th Chinese National Conference on Computational Linguistics
Month:
October
Year:
2020
Address:
Haikou, China
Venue:
CCL
SIG:
Publisher:
Chinese Information Processing Society of China
Note:
Pages:
616–623
Language:
Chinese
URL:
https://www.aclweb.org/anthology/2020.ccl-1.57
DOI:
Bib Export formats:
BibTeX MODS XML EndNote
PDF:
http://aclanthology.lst.uni-saarland.de/2020.ccl-1.57.pdf