Structured Sentiment Analysis as Dependency Graph Parsing

TL;DR

This paper introduces a unified dependency graph parsing approach for structured sentiment analysis, improving extraction of opinion tuples across multiple languages and datasets by leveraging syntactic information.

Contribution

It proposes a novel framework that treats sentiment analysis as dependency graph parsing, unifying sub-tasks and enhancing performance over existing methods.

Findings

Significant performance improvements over state-of-the-art baselines.

Effective across five datasets in four languages.

Refining sentiment graphs with syntactic dependencies further boosts results.

Abstract

Structured sentiment analysis attempts to extract full opinion tuples from a text, but over time this task has been subdivided into smaller and smaller sub-tasks, e,g,, target extraction or targeted polarity classification. We argue that this division has become counterproductive and propose a new unified framework to remedy the situation. We cast the structured sentiment problem as dependency graph parsing, where the nodes are spans of sentiment holders, targets and expressions, and the arcs are the relations between them. We perform experiments on five datasets in four languages (English, Norwegian, Basque, and Catalan) and show that this approach leads to strong improvements over state-of-the-art baselines. Our analysis shows that refining the sentiment graphs with syntactic dependency information further improves results.