摘要
方面级情感分析(ABSA)任务旨在判断评论语句中特定方面词的情感极性。在ABSA领域中,同时提取语法和语义这2种信息的双通道模型取得了一定的效果。然而,现有模型未能考虑语法节点间的重要程度不同、全局范围下的注意力机制引入的额外噪声以及同类特征间存在一定关联性等问题。为了解决以上问题,提出一种依赖类型及距离增强的双通道图卷积模型。首先,在语法模块引入依赖类型以衡量不同邻近节点的重要程度;其次,以依赖树距离为依据构造掩码矩阵进而过滤与语法无关的噪声;最后,引入一个有监督对比损失帮助模型学习同类特征间的关联性。实验结果表明,相较于次优模型DGNN(Dual Graph Neural Network),所提模型在SemEval-2014 Restaurant、SemEval-2014 Laptop和Twitter这3个数据集上分别取得了0.11、0.94和1.01个百分点的准确率提升,以及0.63、1.66和0.83个百分点的宏F1值提升,验证了所提模型的有效性。
Aspect-Based Sentiment Analysis(ABSA)tasks aim to determine the sentiment polarity of specific aspect words in comments.In the field of ABSA,dual-channel models that extract both grammar and semantic information have achieved certain results.However,the existing models fail to consider the different degrees of importance among grammar nodes,the additional noise introduced by global attention mechanism,and the existence of correlations between similar features comprehensively.To address these issues,a dual-channel graph convolutional model with dependency type and distance enhancements was proposed.Firstly,dependency types were introduced in the grammar module to measure the importance of neighborhood nodes.Secondly,mask matrices based on the dependency tree distance were constructed to filter out grammar unrelated noise.Finally,a supervised contrastive loss was introduced to facilitate the model to learn correlations between similar features.Experimental results show that on SemEval-2014 Restaurant,SemEval-2014 Laptop and Twitter datasets,compared to the second-best model DGNN(Dual Graph Neural Network),the proposed model achieves accuracy improvements of 0.11,0.94,1.01 percentage points,respectively,and Macro-F1 improvements of 0.63,1.66,0.83 percentage points,respectively,verifying the effectiveness of the proposed model.
作者
赵彪
秦玉华
田荣坤
胡月航
陈芳锐
ZHAO Biao;QIN Yuhua;TIAN Rongkun;HU Yuehang;CHEN Fangrui(School of Information Science and Technology,Qingdao University of Science and Technology,Qingdao Shandong 266061,China;Technical Research Center,China Tobacco Yunnan Industrial Company Limited,Kunming Yunnan 650024,China)
出处
《计算机应用》
北大核心
2025年第8期2507-2514,共8页
journal of Computer Applications
基金
青岛市科技惠民示范项目(23-2-8-smjk-20-nsh)。
关键词
方面级别情感分析
图神经网络
依赖类型
依赖树距离
有监督对比损失
Aspect-Based Sentiment Analysis(ABSA)
graph convolutional network
dependency type
dependency tree distance
supervised contrastive loss
