From Graphs to Hypergraphs: Enhancing Aspect-Based Sentiment Analysis via Multi-Level Relational Modeling

TL;DR

This paper introduces HyperABSA, a dynamic hypergraph framework that improves aspect-based sentiment analysis by modeling complex relations more effectively than traditional graph-based methods, especially in short texts.

Contribution

The paper proposes a novel dynamic hypergraph approach with an adaptive clustering method, outperforming existing graph-based models in ABSA tasks and reducing redundancy and error propagation.

Findings

HyperABSA achieves consistent improvements on benchmark datasets.

The hypergraph approach outperforms traditional graph models, especially with RoBERTa.

Adaptive clustering enhances hyperedge construction efficiency.

Abstract

Aspect-Based Sentiment Analysis (ABSA) predicts sentiment polarity for specific aspect terms, a task made difficult by conflicting sentiments across aspects and the sparse context of short texts. Prior graph-based approaches model only pairwise dependencies, forcing them to construct multiple graphs for different relational views. These introduce redundancy, parameter overhead, and error propagation during fusion, limiting robustness in short-text, low-resource settings. We present HyperABSA, a dynamic hypergraph framework that induces aspect-opinion structures through sample-specific hierarchical clustering. To construct these hyperedges, we introduce a novel acceleration-fallback cutoff for hierarchical clustering, which adaptively determines the level of granularity. Experiments on three benchmarks (Lap14, Rest14, MAMS) show consistent improvements over strong graph baselines, with substantial gains when paired with RoBERTa backbones. These results position dynamic hypergraph construction as an efficient, powerful alternative for ABSA, with potential extensions to other short-text NLP tasks.