Open-Domain Targeted Sentiment Analysis via Span-Based Extraction and Classification
This work addresses sentiment analysis for applications needing precise opinion target detection, though it is incremental as it builds on existing span-based approaches.
The paper tackles open-domain targeted sentiment analysis by proposing a span-based extract-then-classify framework to address issues like huge search space and sentiment inconsistency in prior sequence tagging methods, achieving consistent performance improvements over baselines on three benchmark datasets.
Open-domain targeted sentiment analysis aims to detect opinion targets along with their sentiment polarities from a sentence. Prior work typically formulates this task as a sequence tagging problem. However, such formulation suffers from problems such as huge search space and sentiment inconsistency. To address these problems, we propose a span-based extract-then-classify framework, where multiple opinion targets are directly extracted from the sentence under the supervision of target span boundaries, and corresponding polarities are then classified using their span representations. We further investigate three approaches under this framework, namely the pipeline, joint, and collapsed models. Experiments on three benchmark datasets show that our approach consistently outperforms the sequence tagging baseline. Moreover, we find that the pipeline model achieves the best performance compared with the other two models.