Revisiting Structured Sentiment Analysis as Latent Dependency Graph Parsing

TL;DR

This paper redefines Structured Sentiment Analysis as a dependency parsing problem on latent trees, explicitly modeling internal span structures, leading to significant performance improvements over previous methods.

Contribution

It introduces a novel two-stage parsing approach with TreeCRFs and a constrained inside algorithm to explicitly model internal span structures in SSA.

Findings

Achieves state-of-the-art results on five benchmark datasets.

Outperforms all previous bi-lexical SSA methods.

Effectively models internal span structures for better sentiment analysis.

Abstract

Structured Sentiment Analysis (SSA) was cast as a problem of bi-lexical dependency graph parsing by prior studies. Multiple formulations have been proposed to construct the graph, which share several intrinsic drawbacks: (1) The internal structures of spans are neglected, thus only the boundary tokens of spans are used for relation prediction and span recognition, thus hindering the model's expressiveness; (2) Long spans occupy a significant proportion in the SSA datasets, which further exacerbates the problem of internal structure neglect. In this paper, we treat the SSA task as a dependency parsing task on partially-observed dependency trees, regarding flat spans without determined tree annotations as latent subtrees to consider internal structures of spans. We propose a two-stage parsing method and leverage TreeCRFs with a novel constrained inside algorithm to model latent structures explicitly, which also takes advantages of joint scoring graph arcs and headed spans for global optimization and inference. Results of extensive experiments on five benchmark datasets reveal that our method performs significantly better than all previous bi-lexical methods, achieving new state-of-the-art.