Transformer-based Text Classification on Unified Bangla Multi-class Emotion Corpus

TL;DR

This paper presents a transformer-based approach for classifying emotions in Bangla texts across six categories, utilizing a newly created unified dataset to achieve high accuracy in emotion detection.

Contribution

It introduces a comprehensive Bangla emotion classifier using transformer models and a new unified dataset, UBMEC, for multi-class emotion recognition in Bangla language.

Findings

High accuracy in emotion classification on UBMEC

Creation of a publicly available Bangla emotion dataset

Effective transformer-based model implementation for Bangla emotions

Abstract

In this research, we propose a complete set of approaches for identifying and extracting emotions from Bangla texts. We provide a Bangla emotion classifier for six classes: anger, disgust, fear, joy, sadness, and surprise, from Bangla words using transformer-based models, which exhibit phenomenal results in recent days, especially for high-resource languages. The Unified Bangla Multi-class Emotion Corpus (UBMEC) is used to assess the performance of our models. UBMEC is created by combining two previously released manually labeled datasets of Bangla comments on six emotion classes with fresh manually labeled Bangla comments created by us. The corpus dataset and code we used in this work are publicly available.