To be Closer: Learning to Link up Aspects with Opinions

TL;DR

This paper introduces an adaptive, aspect-centric tree structure for ABSA that improves the proximity of aspect and opinion words, leading to better sentiment polarity detection.

Contribution

It proposes a novel learnable tree structure that dynamically shortens aspect-opinion distances, enhancing ABSA performance over static dependency trees.

Findings

Significant improvement over strong baselines.

Average aspect-opinion distance reduced by at least 19%.

Effective on five different datasets.

Abstract

Dependency parse trees are helpful for discovering the opinion words in aspect-based sentiment analysis (ABSA). However, the trees obtained from off-the-shelf dependency parsers are static, and could be sub-optimal in ABSA. This is because the syntactic trees are not designed for capturing the interactions between opinion words and aspect words. In this work, we aim to shorten the distance between aspects and corresponding opinion words by learning an aspect-centric tree structure. The aspect and opinion words are expected to be closer along such tree structure compared to the standard dependency parse tree. The learning process allows the tree structure to adaptively correlate the aspect and opinion words, enabling us to better identify the polarity in the ABSA task. We conduct experiments on five aspect-based sentiment datasets, and the proposed model significantly outperforms recent strong baselines. Furthermore, our thorough analysis demonstrates the average distance between aspect and opinion words are shortened by at least 19% on the standard SemEval Restaurant14 dataset.