Emotion Classification in a Resource Constrained Language Using Transformer-based Approach

TL;DR

This paper introduces a transformer-based approach for emotion classification in Bengali, a resource-constrained language, developing a new dataset and demonstrating XLM-R's superior performance over other models.

Contribution

It presents a new Bengali emotion dataset and evaluates multiple models, highlighting the effectiveness of transformer-based methods like XLM-R for emotion classification in low-resource languages.

Findings

XLM-R achieved the highest weighted F1-score of 69.73%.

The dataset is publicly available for future research.

Transformer models outperform traditional machine learning and neural network approaches.

Abstract

Although research on emotion classification has significantly progressed in high-resource languages, it is still infancy for resource-constrained languages like Bengali. However, unavailability of necessary language processing tools and deficiency of benchmark corpora makes the emotion classification task in Bengali more challenging and complicated. This work proposes a transformer-based technique to classify the Bengali text into one of the six basic emotions: anger, fear, disgust, sadness, joy, and surprise. A Bengali emotion corpus consists of 6243 texts is developed for the classification task. Experimentation carried out using various machine learning (LR, RF, MNB, SVM), deep neural networks (CNN, BiLSTM, CNN+BiLSTM) and transformer (Bangla-BERT, m-BERT, XLM-R) based approaches. Experimental outcomes indicate that XLM-R outdoes all other techniques by achieving the highest weighted $f_1$-score of $69.73\%$ on the test data. The dataset is publicly available at https://github.com/omar-sharif03/NAACL-SRW-2021.