Towards Label-Agnostic Emotion Embeddings

TL;DR

This paper introduces a training scheme that creates universal emotion embeddings, enabling cross-format, cross-language, and cross-genre emotion analysis without sacrificing accuracy.

Contribution

It proposes a novel training approach for shared emotion representations that are independent of label formats, languages, and model architectures.

Findings

Achieves interoperability across diverse emotion datasets

Maintains high prediction quality across formats and languages

Provides open-source code and data for reproducibility

Abstract

Research in emotion analysis is scattered across different label formats (e.g., polarity types, basic emotion categories, and affective dimensions), linguistic levels (word vs. sentence vs. discourse), and, of course, (few well-resourced but much more under-resourced) natural languages and text genres (e.g., product reviews, tweets, news). The resulting heterogeneity makes data and software developed under these conflicting constraints hard to compare and challenging to integrate. To resolve this unsatisfactory state of affairs we here propose a training scheme that learns a shared latent representation of emotion independent from different label formats, natural languages, and even disparate model architectures. Experiments on a wide range of datasets indicate that this approach yields the desired interoperability without penalizing prediction quality. Code and data are archived under DOI 10.5281/zenodo.5466068.