From Graphs to Hypergraphs: Enhancing Aspect-Based Sentiment Analysis via Multi-Level Relational Modeling
This work solves the challenge of modeling multi-level dependencies in ABSA for NLP researchers and practitioners, offering an efficient alternative that could extend to other short-text tasks, though it is incremental as it builds on existing graph-based approaches.
The paper tackled the problem of Aspect-Based Sentiment Analysis (ABSA) by addressing limitations of prior graph-based methods, such as redundancy and error propagation, through a dynamic hypergraph framework called HyperABSA, which achieved consistent improvements on benchmarks like Lap14, Rest14, and MAMS, with substantial gains when using RoBERTa backbones.
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.