Key points are not available for this paper at this time.
There have been growing concerns about the influence of hate speech on social discourse and its ability to instigate violence and prejudice as it has spread widely across internet platforms. Researchers and service providers must now prioritize identifying and regulating hate speech. In this survey, we look at studies published between 2018 and 2023 that explore various aspects of hate speech identification. This review begins by pointing out the alarming growth of hate speech on the internet and its adverse effects, underscoring the importance of developing reliable identification mechanisms. Based on the papers' principal focuses, we classify them into one of five broad themes: dataset construction, algorithm development, bias analysis, multilingual and multimodal techniques, and ethical considerations. This systematic literature review and meta-analysis highlights the need for standardized evaluation metrics, more extensive datasets, and robust algorithms to deal with the ever-evolving nature of hate speech while pointing out the shortcomings of currently available hate speech identification methods and datasets. To effectively counteract online hate speech, researchers, legislators, and technology businesses will find this comprehensive assessment an invaluable resource, an in-depth overview of the hate speech identification landscape. Future research initiatives on this crucial topic can build upon the insights and problems given here.
Saleem et al. (Wed,) studied this question.