If you've got it, flaunt it: Making the most of fine-grained sentiment annotations

TL;DR

This paper investigates how incorporating holder and expression annotations can enhance fine-grained sentiment analysis, demonstrating improvements in target extraction and polarity classification, while highlighting current challenges in predicting polar expressions.

Contribution

It introduces methods to jointly predict target and polarity labels and shows that adding gold expression annotations improves classification performance.

Findings

Joint target and polarity prediction enhances extraction accuracy.

Adding gold expressions improves targeted polarity classification.

Current models struggle to accurately predict polar expressions.

Abstract

Fine-grained sentiment analysis attempts to extract sentiment holders, targets and polar expressions and resolve the relationship between them, but progress has been hampered by the difficulty of annotation. Targeted sentiment analysis, on the other hand, is a more narrow task, focusing on extracting sentiment targets and classifying their polarity.In this paper, we explore whether incorporating holder and expression information can improve target extraction and classification and perform experiments on eight English datasets. We conclude that jointly predicting target and polarity BIO labels improves target extraction, and that augmenting the input text with gold expressions generally improves targeted polarity classification. This highlights the potential importance of annotating expressions for fine-grained sentiment datasets. At the same time, our results show that performance of current models for predicting polar expressions is poor, hampering the benefit of this information in practice.