Comprehensive Analysis of the Artificial Intelligence Approaches for Detecting Misogynistic Mixed-Code Online Content in South Asian Countries

Abstract

The rise of social media has drastically altered several aspects of daily life and businesses. With all its advantages, the anonymity and lack of accountability social media provides encourages unsavoury individuals to spread hate. Hate targeted towards a particular group, such as women, can have a silencing effect and discourage them from participating in online discourse. In this chapter, the authors review recent studies and toolkits that attempt to tackle the issue of hate speech on online platforms using natural language processing (NLP) techniques. Challenges and shared tasks that are regularly conducted to advance the current state-of-the-art in hate speech detection in English and other under-resourced languages are also reviewed. The comprehensive survey suggests that despite the recent increase in interest in the problem of filtering online hate speech, the field is still in its infancy, specifically the problem of misogyny identification in under-resourced languages.