摘要
语义角色标记被认为是实现自然语言理解的关键一步,并已被广泛研究.近年来,基于深度神经网络的端到端语义角色标记越来越受到关注.然而当前的语义角色标记方法使用没有语言特征的深度神经网络,配价结构可以较好地刻画句子的句法结构和语义构成关系,因此,本文提出了一种融合配价信息的深度神经网络模型,它有效地结合了丰富的语言配价信息以进行语义角色标记.本文在CoNLL-2005共享任务数据集和CoNLL-2012共享任务数据集上对融合配价信息的深度神经网络模型进行了评估,在两个基准语义角色标记数据集上比以前的工作取得了更好的结果.
Semantic roles labeling is considered to be a crucial step towards natural language understanding and has been widely studied.In recent years,end-to-end semantic role labeling based on deep neural networks has attracted more and more attention.However,current semantic role labeling methods use deep neural networks without language features.Valence structure can better describe the syntactic structure and semantic structure of sentences.Therefore,this paper proposes a deep neural network model that integrates valence information,which effectively combines rich language valence information for semantic role labeling.It was evaluated on the CoNLL-2005 shared task dataset and the CoNLL-2012 shared task dataset,and achieved better results than previous work on the two benchmark semantic role labeling datasets.
作者
袁里驰
YUAN Li-chi(School of Software and Internet of Things Engineering,Jiangxi University of Finance and Economics,Nanchang 330013,China)
出处
《小型微型计算机系统》
CSCD
北大核心
2022年第9期1925-1930,共6页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61962025,61562034)资助.
关键词
深度神经网络
语义角色标注
配价结构
deep neural network
semantic role labeling
valence structure
