Evaluating Emotion Arcs Across Languages: Bridging the Global Divide in Sentiment Analysis

TL;DR

This study systematically evaluates automatic emotion arc generation methods across multiple languages, demonstrating that lexicon-based approaches are effective at aggregate levels and that translation can extend these methods to low-resource languages, advancing global emotion analysis.

Contribution

First comprehensive quantitative evaluation of emotion arc generation methods across diverse languages, highlighting the effectiveness of lexicon-based approaches and translation for low-resource languages.

Findings

Lexicon-only methods are highly accurate at aggregate emotion arc generation.

Translation of English lexicons enables emotion analysis in less-resource languages.

Lexicon-based approaches outperform machine learning at the arc level despite poor instance-level classification.

Abstract

Emotion arcs capture how an individual (or a population) feels over time. They are widely used in industry and research; however, there is little work on evaluating the automatically generated arcs. This is because of the difficulty of establishing the true (gold) emotion arc. Our work, for the first time, systematically and quantitatively evaluates automatically generated emotion arcs. We also compare two common ways of generating emotion arcs: Machine-Learning (ML) models and Lexicon-Only (LexO) methods. By running experiments on 18 diverse datasets in 9 languages, we show that despite being markedly poor at instance level emotion classification, LexO methods are highly accurate at generating emotion arcs when aggregating information from hundreds of instances. We also show, through experiments on six indigenous African languages, as well as Arabic, and Spanish, that automatic translations of English emotion lexicons can be used to generate high-quality emotion arcs in less-resource languages. This opens up avenues for work on emotions in languages from around the world; which is crucial for commerce, public policy, and health research in service of speakers often left behind. Code and resources: https://github.com/dteodore/EmotionArcs