Abstract

Social media platforms offer avenues for fostering anonymous online connections, discussions on diverse topics like
culture, politics, and community life. However, the proliferation of hate speech poses a pressing challenge for
society, individuals, policymakers, and researchers alike, both on the continent and globally. Addressing this issue
necessitates comprehensive studies to identify and combat hate speech effectively. This paper conducts a systematic
review of literature in this domain, concentrating on methodologies such as word embedding, machine learning,
deep learning, and cutting-edge technologies. Specifically focusing on the past six years of research, this review
highlights gaps, challenges, and advancements in hate speech detection techniques. Additionally, it delves into
limitations, algorithmic selection dilemmas, data collection complexities, cleaning challenges, and outlines future
research pathways in this critical area.
Keywords: Hate Speech Detection, Machine Learning, Social Media Platforms, Text Analysis, Algorithm
Selection.