Understanding the Role of Affect Dimensions in Detecting Emotions from Tweets: A Multi-task Approach

TL;DR

This paper introduces VADEC, a multi-task framework that jointly models emotion classification and dimensional regression from tweets, significantly improving emotion detection accuracy and establishing state-of-the-art results across multiple datasets.

Contribution

The paper presents VADEC, a novel multi-task learning approach that leverages the correlation between categorical and dimensional emotion models for improved tweet emotion detection.

Findings

Outperforms baselines with 3.4%, 11%, and 3.9% gains on AIT dataset

Achieves 11.3% average gains over six metrics on SenWave dataset

Improves Pearson Correlation scores by 7.6% and 16.5% for Valence and Dominance dimensions

Abstract

We propose VADEC, a multi-task framework that exploits the correlation between the categorical and dimensional models of emotion representation for better subjectivity analysis. Focusing primarily on the effective detection of emotions from tweets, we jointly train multi-label emotion classification and multi-dimensional emotion regression, thereby utilizing the inter-relatedness between the tasks. Co-training especially helps in improving the performance of the classification task as we outperform the strongest baselines with 3.4%, 11%, and 3.9% gains in Jaccard Accuracy, Macro-F1, and Micro-F1 scores respectively on the AIT dataset. We also achieve state-of-the-art results with 11.3% gains averaged over six different metrics on the SenWave dataset. For the regression task, VADEC, when trained with SenWave, achieves 7.6% and 16.5% gains in Pearson Correlation scores over the current state-of-the-art on the EMOBANK dataset for the Valence (V) and Dominance (D) affect dimensions respectively. We conclude our work with a case study on COVID-19 tweets posted by Indians that further helps in establishing the efficacy of our proposed solution.