知识驱动的多语义语法交互方面级情感分析  

Knowledge-driven Multi Semantic Syntactic Interactions Model for Aspect-based Sentiment Analysis

作　　者：白雪 韩虎[1] BAI Xue;HAN Hu(School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)

机构地区：[1]兰州交通大学电子与信息工程学院,兰州730070

出　　处：《兰州交通大学学报》2024年第3期49-56,67,共9页Journal of Lanzhou Jiaotong University

基　　金：国家自然科学基金(62166024)。

摘　　要：针对目前图卷积网络处理情感分类任务时存在对文本丰富的信息利用不充分问题,提出一种知识驱动的多语义语法交互方面级情感分析模型。通过方面词与上下文之间的多种语义关系构建多语义关系图,利用外部知识构建情感依存图;此外,通过对注意力机制进行优化获得更高级上下文表示,随后融合不同表示。5个数据集上的实验结果表明:所提模型与基准模型ASGCN相比,KSSGCN_GloVe的准确率和宏F1分别平均提升1.59%和3.17%。In response to the problem of insufficient utilization of rich context information in sentiment classification tasks using graph convolutional networks,a knowledge-driven aspect based sentiment analysis model for multi-se-mantic syntactic interaction is proposed.Multiple semantic relationships between aspects and contexts are used to construct a multi-semantic relationship diagram,and external knowledge is used to construct an emotional dependen-cy diagram,In addition,a more advanced context representation is obtained by optimizing the attention mechanism.Then,different representations were fused.The experimental results on five public datasets show that the proposed model,KSSGCN_GloVe,has achieved an average improvement of 1.59%in accuracy and 3.17%in macro F1 com-pared to the baseline model ASGCN.

关 键 词：方面级情感分析 知识驱动 图卷积网络 外部知识 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]