BongHope: An annotated corpus for Bengali hope speech detection

Nath, Tanusree, Singh, Vivek Kumar and Gupta, Vedika (2025) BongHope: An annotated corpus for Bengali hope speech detection. International Journal of Information Technology. ISSN 2511-2104 (In Press)

[thumbnail of s41870-025-02484-2.pdf] Text
s41870-025-02484-2.pdf - Published Version
Restricted to Repository staff only

Download (685kB) | Request a copy

Abstract

The exponential growth of social media has fostered the spread of both negativity (hate speech) and supportive content, with the latter often categorized as hope speech—text promoting peace and a hopeful outlook. Recently, a few research works have been conducted on automatic detection of hope speech in different languages, including English, Tamil, Malayalam, Spanish and Kannada. However, to the best of our knowledge, there is no research on hope speech in Bengali language text. Despite Bengali’s significant presence on social media, hope speech detection in this language remains unexplored. Therefore, it is important to develop appropriate computational methods for the automatic detection of hope and non-hope speech in Bengali text. One possible reason for the lack of hope speech research in Bengali may be the unavailability of a suitable dataset or corpus for this purpose. This study presents the first curated and annotated dataset for hope speech in Bengali, enabling effective detection and analysis. Several state-of-the-art computational models are applied to the created dataset and the results obtained confirm the suitability of the dataset for hope speech research. Overall, this research provides a foundational resource for Bengali hope speech detection, contributing to multilingual social media analysis.

Item Type: Article
Keywords: Bengali language | Hope speech | Hope speech corpus | Hope speech dataset | Social media data analytics
Subjects: Physical, Life and Health Sciences > Computer Science
Social Sciences and humanities > Social Sciences > Social Sciences (General)
Social Sciences and humanities > Social Sciences > Journalism, News and Media
JGU School/Centre: Jindal Global Business School
Depositing User: Dharmveer Modi
Date Deposited: 01 Apr 2025 13:05
Last Modified: 02 Apr 2025 09:19
Official URL: https://doi.org/10.1007/s41870-025-02484-2
URI: https://pure.jgu.edu.in/id/eprint/9334

Downloads

Downloads per month over past year

Actions (login required)

View Item
View Item