Multi-View Attention Syntactic Enhanced Graph Convolutional Network for Aspect-based Sentiment Analysis

TL;DR

MASGCN addresses ABSA by jointly leveraging diverse syntactic views through distance-based masks and a dependency-type matrix learned with structural entropy, then fusing these views with a learned multi-view attention mechanism. The model introduces a structural-entropy loss to exploit dependency-type information and combines semantic and syntactic cues within a multi-view GNN framework, achieving state-of-the-art results on four benchmarks. Empirical results show substantial gains over strong baselines, with ablation and hyperparameter analyses confirming the contribution of each component. The approach offers a scalable path to richer syntactic information utilization in ABSA and has potential for integration with external knowledge sources in future work.

Abstract

Aspect-based Sentiment Analysis (ABSA) is the task aimed at predicting the sentiment polarity of aspect words within sentences. Recently, incorporating graph neural networks (GNNs) to capture additional syntactic structure information in the dependency tree derived from syntactic dependency parsing has been proven to be an effective paradigm for boosting ABSA. Despite GNNs enhancing model capability by fusing more types of information, most works only utilize a single topology view of the dependency tree or simply conflate different perspectives of information without distinction, which limits the model performance. To address these challenges, in this paper, we propose a new multi-view attention syntactic enhanced graph convolutional network (MASGCN) that weighs different syntactic information of views using attention mechanisms. Specifically, we first construct distance mask matrices from the dependency tree to obtain multiple subgraph views for GNNs. To aggregate features from different views, we propose a multi-view attention mechanism to calculate the attention weights of views. Furthermore, to incorporate more syntactic information, we fuse the dependency type information matrix into the adjacency matrices and present a structural entropy loss to learn the dependency type adjacency matrix. Comprehensive experiments on four benchmark datasets demonstrate that our model outperforms state-of-the-art methods. The codes and datasets are available at https://github.com/SELGroup/MASGCN.

Figures (3)

• Figure 1: Overall framework of MASGCN.
• Figure 2: Hyperparameter sensitivity analysis.
• Figure 3: Insights of matrices.