Learn from Structural Scope: Improving Aspect-Level Sentiment Analysis with Hybrid Graph Convolutional Networks

TL;DR

This paper proposes a hybrid graph convolutional network that leverages syntactic structures to improve aspect-level sentiment analysis by modeling the relation between targets and sentiments more effectively.

Contribution

It introduces the concept of Scope and combines constituency and dependency trees using HGCN for enhanced sentiment analysis accuracy.

Findings

Outperforms state-of-the-art baselines on four datasets

Effectively models syntactic relations for sentiment detection

Enriches representations by linking two syntax parsing methods

Abstract

Aspect-level sentiment analysis aims to determine the sentiment polarity towards a specific target in a sentence. The main challenge of this task is to effectively model the relation between targets and sentiments so as to filter out noisy opinion words from irrelevant targets. Most recent efforts capture relations through target-sentiment pairs or opinion spans from a word-level or phrase-level perspective. Based on the observation that targets and sentiments essentially establish relations following the grammatical hierarchy of phrase-clause-sentence structure, it is hopeful to exploit comprehensive syntactic information for better guiding the learning process. Therefore, we introduce the concept of Scope, which outlines a structural text region related to a specific target. To jointly learn structural Scope and predict the sentiment polarity, we propose a hybrid graph convolutional network (HGCN) to synthesize information from constituency tree and dependency tree, exploring the potential of linking two syntax parsing methods to enrich the representation. Experimental results on four public datasets illustrate that our HGCN model outperforms current state-of-the-art baselines.