Work

A Comprehensive Survey on Urdu Hate Speech Detection: Methods, Evaluation, and Challenges

Year: 2025

Type: article

Abstract: Hate speeches has risen significantly because of ever expanding online communities, easy accessibility of social media platforms, anonymity of the users, rapid spread of information, and increased visibility of hate related incidents. Detection of such hate speech, especially in low-resource language i.e. Urdu, on social media has become active research area for safe, and inclusive online environment. This paper introduces a new taxonomy, and presents a survey of Urdu hate speech detection techniques. It also explores future research directions, and highlights the main challenges which persist in this field such as limited data availability, code-switching, and requirements for adaptability to contextual perspectives. The taxonomy, and the survey form the basis for continued advancement in Urdu hate speech detection that yields improved insights regarding Urdu hate speech patterns, and contributes to better detection system’s development. It serves a comprehensive review of Urdu hate speech detection methods available in the research literature, and compares those methods at an abstract level, discusses their limitations, and provides future research directions for new researches in this domain.

Source: IEEE Access

DOI: 10.1109/access.2025.3591143

Authors: Ijaz Hussain, Umar Farooq, Ammara Nawaz Cheema, Ibrahim M. Almanjahie

Institutions: Pakistan Institute of Engineering and Applied Sciences, Air University, King Khalid University

Language: English

Topic: Hate Speech and Cyberbullying Detection

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Reduced inequalities

Open Access status: gold

Estimated APC paid: $2,075