基于多尺度混合注意力卷积神经网络的关系抽取模型  

Relation extraction model based on multi-scale hybrid attention convolutional neural networks

作　　者：唐媛 陈艳平 扈应 黄瑞章[1,2,3] 秦永彬[1,2,3] TANG Yuan;CHEN Yanping;HU Ying;HUANG Ruizhang;QIN Yongbin(Text Computing and Cognitive Intelligence Engineering Research Center of Ministry of Education,Guizhou University,Guiyang Guizhou 550025,China;State Key Laboratory of Public Big Data(Guizhou University),Guiyang Guizhou 550025,China;College of Computer Science and Technology,Guizhou University,Guiyang Guizhou 550025,China)

机构地区：[1]贵州大学文本计算与认知智能教育部工程研究中心,贵阳550025 [2]公共大数据国家重点实验室(贵州大学),贵阳550025 [3]贵州大学计算机科学与技术学院,贵阳550025

出　　处：《计算机应用》2024年第7期2011-2017,共7页journal of Computer Applications

基　　金：国家自然科学基金资助项目(62166007);贵州省科技支撑计划项目([2022]277)。

摘　　要：针对基于卷积神经网络(CNN)的关系抽取获取句子语义信息时缺少不同尺度语义特征信息的获取以及对关键信息的关注的问题,提出基于多尺度混合注意力CNN的关系抽取模型。首先,将关系抽取建模为二维化表示的标签预测;其次,通过多尺度的特征信息提取与融合,获得更细粒度的多尺度空间信息;然后,通过注意力与卷积的结合自适应地细化特征图,使模型关注重要的上下文信息;最后,使用两个预测器共同预测实体对之间的关系标签。实验结果表明,多尺度混合卷积注意力模型能够获取多尺度语义特征信息,而通道注意力和空间注意力通过权重捕捉通道和空间的关键信息,以此来提升关系抽取的性能。所提模型在数据集SemEval(SemEval-2010 task 8)、TACRED(TAC Relation Extraction Dataset)、Re-TACRED(Revised-TACRED)和SciERC(Entities,Relations,and Coreference for Scientific knowledge graph construction)上的F1值分别达到90.32%、70.74%、85.71%和89.66%。To address the issue of insufficient extraction of semantic feature information with different scales and the lack of focus on crucial information when obtaining sentence semantic information by Convolutional Neural Network(CNN)-based relation extraction,a model for relation extraction based on a multi-scale hybrid attention CNN was proposed.Firstly,relation extraction was modeled as label prediction with two-dimensional representation.Secondly,by extracting and fusing multi-scale feature information,finer-grained multi-scale spatial information was obtained.Thirdly,through the combination of attention and convolution,the feature maps were refined adaptively to make the model concentrate on important contextual information.Finally,two predictors were used jointly to predict the relation labels between entity pairs.Experimental results demonstrate that the multi-scale hybrid convolutional attention model can capture multi-scale semantic feature information,And the key information in channels and spatial locations was captured by the channel attention and spatial attention by assigning appropriate weights,thereby improving the performance of relation extraction.The proposed model achieves F1 scores of 90.32%on SemEval(SemEval-2010 task 8)dataset,70.74%on TACRED(TAC Relation Extraction Dataset),85.71%on Re-TACRED(Revised-TACRED),and 89.66%on SciERC(Entities,Relations,and Coreference for Scientific knowledge graph construction).

关 键 词：关系抽取 二维化表示 通道注意力 空间注意力 多尺度语义特征 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]