A Comparative Analysis of Noise Reduction Methods in Sentiment Analysis on Noisy Bangla Texts

TL;DR

This paper introduces a new dataset and evaluates noise reduction methods for sentiment analysis on noisy Bangla texts, revealing current methods are inadequate and highlighting the need for improved techniques.

Contribution

It presents a manually annotated dataset of noisy Bangla texts, formulates noise identification as a multi-label classification task, and benchmarks baseline noise reduction methods for sentiment analysis.

Findings

Baseline noise reduction methods are ineffective on noisy Bangla texts.

Noise reduction methods do not significantly improve sentiment analysis accuracy.

The study provides publicly available dataset and implementation for future research.

Abstract

While Bangla is considered a language with limited resources, sentiment analysis has been a subject of extensive research in the literature. Nevertheless, there is a scarcity of exploration into sentiment analysis specifically in the realm of noisy Bangla texts. In this paper, we introduce a dataset (NC-SentNoB) that we annotated manually to identify ten different types of noise found in a pre-existing sentiment analysis dataset comprising of around 15K noisy Bangla texts. At first, given an input noisy text, we identify the noise type, addressing this as a multi-label classification task. Then, we introduce baseline noise reduction methods to alleviate noise prior to conducting sentiment analysis. Finally, we assess the performance of fine-tuned sentiment analysis models with both noisy and noise-reduced texts to make comparisons. The experimental findings indicate that the noise reduction methods utilized are not satisfactory, highlighting the need for more suitable noise reduction methods in future research endeavors. We have made the implementation and dataset presented in this paper publicly available at https://github.com/ktoufiquee/A-Comparative-Analysis-of-Noise-Reduction-Methods-in-Sentiment-Analysis-on-Noisy-Bangla-Texts