CUET_Zahra_Duo@Mahed 2025: Hate and Hope Speech Detection in Arabic Social Media Content using Transformer
Year: 2025
Type: conference-paper
Abstract: In recent years, online social life has become an integral part of the global landscape, with social media platforms enabling users to express a wide range of emotions and opinions.In the Arabic-speaking world, navigating the dual nature of content-encompassing both hate and hope speech-remains challenging due to linguistic and cultural complexities.The MAHED 2025 shared task at ArabicNLP 2025 addressed this by focusing on detecting both hate and hope speech in Arabic social media.This paper describes our approach for subtask 1, utilizing various machine learning, deep learning, and transformer models for classification.AraBERT-large-v2 yielded the highest macro f 1 -score of 0.698, earning 8 th place on the leaderboard.
Authors: Walisa Alam, Mehreen Rahman, Shawly Ahsan, Mohammed Moshiul Hoque
Institution: Chittagong University of Engineering & Technology
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4416037607