• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

计算机工程与科学 ›› 2023, Vol. 45 ›› Issue (12): 2246-2255.

• 人工智能与数据挖掘 • 上一篇    下一篇

面向多方面的双通道知识增强图卷积网络模型

陈景景1,韩虎2,徐学锋2   

  1. (1.兰州交通大学数理学院,甘肃 兰州 730070;2.兰州交通大学电子与信息工程学院,甘肃 兰州 730070)
  • 收稿日期:2022-08-07 修回日期:2022-09-23 接受日期:2023-12-25 出版日期:2023-12-25 发布日期:2023-12-14
  • 基金资助:
    国家自然科学基金(62166024)

A multi-aspect oriented dual channel knowledge-enhanced graph convolutional network model

CHEN Jing-jing1,HAN Hu2,XU Xue-feng2   

  1.  (1.School of Mathematics and Physics,Lanzhou Jiaotong University,Lanzhou 730070;
    2.School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
  • Received:2022-08-07 Revised:2022-09-23 Accepted:2023-12-25 Online:2023-12-25 Published:2023-12-14

摘要: 基于方面的情感分析是一项细粒度的情感分析任务,旨在将方面与相应的情感词对齐,以进行特定于方面的情感极性推理。近年来,借助句法依赖信息的图神经网络情感分类方法成为该领域的一个研究热点,但是由于评论语句在内容表达和句法结构上的灵活性,仅利用句法依赖信息的建模方法仍然存在一定的不足。为了发挥情感知识与结构语义信息对评论语句的增强作用,提出一种双通道知识增强图卷积网络模型DualSyn-GCN。一方面根据方面与方面、方面与上下文之间的隐含关系进行句法依赖邻接矩阵的增强,另一方面从外部情感知识对方面的情感依赖进行学习,随后对2种不同增强表示进行融合,从而实现不同表示间的共享与互补。实验结果表明,相较于经典的基于特定方面的图卷积网络模型(ASGCN),该模型在LAP14数据集上的准确率和MF1值分别提升了2.34%和3.26%。

关键词: 方面级情感分析, 句法依赖, 知识增强, 图卷积网络, 情感知识

Abstract: Aspect-based sentiment analysis is a fine-grained sentiment analysis task, which aims to align aspects with the corresponding emotion words for aspect specific emotion polarity reasoning. In recent years, the graph neural network sentiment classification method based on syntactic dependent information has become a research hotspot in this field. However, due to the flexibility of comment sentences in content expression and syntactic structure, the modeling method using only syntactic dependent information still has some shortcomings. In order to enhance the comment sentences by affective knowledge and structural semantic information, a convolutional network model(DualSyn-GCN) of two channel knowledge enhancement graph is proposed. On one hand, the syntactic dependency adjacency matrix is enhanced according to the implicit relationship between aspect and aspect as well as aspect and context. On the other hand, the emotional dependency of aspect is learned from external emotional knowledge, and then the two different enhanced representations are fused to realize the sharing and complementarity between different representations. The experimental results show that, compared with the classical aspect based graph convolutional network model (ASGCN), this model improves the accuracy and MF1 value on LAP14 data set by 2.34% and 3.26% respectively. 

Key words: aspect-based sentiment analysis, syntactic dependency, knowledge enhancement, graph convolutional network, affective knowledge