Sentiment Analysis on Bangla Product Reviews with Emojis and Slangs

Published: 15 June 2026| Version 1 | DOI: 10.17632/3k2tdjscvg.1
Contributors:
,
,

Description

This dataset accompanies the paper "Sentiment Analysis on Bangla Product Reviews with Emojis and Slangs" (DOI: 10.1109/ICAIII69475.2026.11521418). The dataset contains raw Bangla product reviews collected from publicly available online sources, including Daraz, Facebook, and Rokomari. The reviews include natural user-generated text containing Bangla language, emojis, informal expressions, and slang terms commonly used in online product feedback. Data were collected, cleaned, and organized to facilitate research in sentiment analysis, natural language processing, and text mining for low-resource languages. This dataset is provided in its raw form and has not been sentiment-labeled. Researchers may use it for data exploration, annotation, preprocessing, feature extraction, sentiment analysis, emoji analysis, slang detection, and other NLP-related tasks. The dataset supports transparency and reproducibility of the associated research while enabling future studies on Bangla-language user-generated content.

Files

Institutions

Categories

Linguistics, Artificial Intelligence, Data Science, Natural Language Processing, Machine Learning

Licence