Exploiting BERT for End-to-End Aspect-based Sentiment Analysis

TL;DR

This paper demonstrates that BERT-based models with simple architectures can outperform state-of-the-art methods in end-to-end aspect-based sentiment analysis, establishing a new benchmark for the task.

Contribution

It introduces a straightforward BERT-based approach for E2E-ABSA and standardizes evaluation practices, serving as a benchmark for future research.

Findings

BERT-based models outperform previous state-of-the-art methods.

Simple linear classification layers with BERT are highly effective.

Standardized evaluation improves comparability of results.

Abstract

In this paper, we investigate the modeling power of contextualized embeddings from pre-trained language models, e.g. BERT, on the E2E-ABSA task. Specifically, we build a series of simple yet insightful neural baselines to deal with E2E-ABSA. The experimental results show that even with a simple linear classification layer, our BERT-based architecture can outperform state-of-the-art works. Besides, we also standardize the comparative study by consistently utilizing a hold-out validation dataset for model selection, which is largely ignored by previous works. Therefore, our work can serve as a BERT-based benchmark for E2E-ABSA.