Work

Deep Multi-Task Models for Misogyny Identification and Categorization on Arabic Social Media

Year: 2022

Type: preprint

Abstract: The prevalence of toxic content on social media platforms, such as hate speech, offensive language, and misogyny, presents serious challenges to our interconnected society. These challenging issues have attracted widespread attention in Natural Language Processing (NLP) community. In this paper, we present the submitted systems to the first Arabic Misogyny Identification shared task. We investigate three multi-task learning models as well as their single-task counterparts. In order to encode the input text, our models rely on the pre-trained MARBERT language model. The overall obtained results show that all our submitted models have achieved the best performances (top three ranked submissions) in both misogyny identification and categorization tasks.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2206.08407

Authors: Mahdaouy, Abdelkader El, Mekki, Abdellah El, Oumar, Ahmed, Mousannif, Hajar, Berrada, Ismail

Institutions: Université Mohammed VI Polytechnique, Cadi Ayyad University

Language: English

Topic: Hate Speech and Cyberbullying Detection

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDGs: Reduced inequalities, Gender equality

Open Access status: green

Funder: Université Mohammed VI Polytechnique

OpenAlex ID: W4283211152

Cites:

Cited by: 2

Related to: 10