结合句法依存信息的方面级情感分类  

Aspect-level sentiment classification combined with syntactic dependency information

作　　者：张沁洋 李旭[1] 姚春龙[1] 李长吾[1] ZHANG Qinyang;LI Xu;YAO Chunlong;LI Changwu(School of Information Science and Engineering,Dalian Polytechnic University,Dalian 116034,Liaoning,China)

机构地区：[1]大连工业大学信息科学与工程学院,辽宁大连116034

出　　处：《山东大学学报（工学版）》2021年第2期83-89,97,共8页Journal of Shandong University（Engineering Science）

基　　金：国家重点研发计划专项(2017YFC0821003-3);辽宁省教育厅科学研究项目(J2020113);辽宁省自然科学基金项目(20180550395)。

摘　　要：引入句法依存信息到原方面术语,提出一种新的方面术语表示方法,利用Glove词向量表示单词以及单词与单词之间的依存关系,构造出包含句法依存信息的依存关系邻接矩阵和依存关系表示矩阵,利用图卷积神经网络和多头注意力机制将句法依存信息融入到方面术语中,使得方面术语表达与上下文结构高度相关。将改进后的方面词术语表示替换到现有模型后,模型泛化能力得到有效提升。对比试验和分析结果表明:该方法具有有效性和泛化性。Considering introducing syntactic dependency information into the original aspect terms, a new aspect term representation method was proposed. First Glove word vector was used to represent the words and dependency relationship between words, and the dependency adjacency matrix and the representation of dependency relationship matrix including syntactic dependency information was constructed. Then graph convolution neural network and multi-head attention mechanism were used to integrate syntactic dependency information into aspect terms, so that aspect terms were highly related to context structure. The models generalization ability were effectively improved by replacing the existing models with the improved aspect term expression. Through comparative experiments and analysis, effectiveness and generalization of the method were proved.

关 键 词：句法依存信息 方面级情感分类 Glove词向量 图卷积 注意力机制 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]