期刊
KNOWLEDGE-BASED SYSTEMS
卷 256, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.knosys.2022.109840
关键词
Graph convolutional networks; Internal semantic correlations; Sentiment interaction relations; Multi-graph perception; Aspect-based sentiment analysis
资金
- National Natural Science Foundation of China [61877050]
- Key R & D plan of Xianyang of China [2021ZDYF-GY-0033]
This paper proposes a novel graph convolutional network with sentiment interaction and multi-graph perception for aspect-based sentiment analysis. The model considers the semantic correlations within aspect phrases and the sentiment interaction relations between different aspects of a sentence. It generates different types of adjacency graphs and uses graph convolutional neural networks and a multi-graph perception mechanism to enrich dependencies and enhance context-awareness. Experimental results show that the proposed model outperforms state-of-the-art methods in terms of accuracy and macro-F1 score.
Graph convolutional networks have been successfully applied to aspect-based sentiment analysis, due to their ability to flexibly capture syntactic information and word dependencies. However, most existing graph network-based models only consider the syntactic dependencies between specific aspects and contexts. These cannot capture the internal semantic correlations within aspect-specific phrases and ignore the sentiment interaction relations between different aspects of a sentence. In this paper, we propose a novel graph convolutional network with sentiment interaction and multi-graph perception for aspect-based sentiment analysis. The proposed model considers the complementarity of semantic dependencies and sentiment interactions simultaneously. Specifically, we generate four types of adjacency graphs by integrating the internal semantic correlations between aspect phrases and linking the sentiment interaction relations among different aspects. Adjacency graphs are used to construct graph convolutional neural networks to enrich aspect-centric dependencies and enhance the capability of context-awareness. In addition, we construct a multi-graph perception mechanism to capture the specific dependency information that cannot be captured between different graphs and hence reduce the amount of overlapping information. Experimental results on five publicly-available datasets demonstrate that our proposed model outperforms state-of-the-art methods and achieves the best performance in terms of accuracy and macro-F1 score. (c) 2022 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据