Comparative Analysis on Detection of Hate Speech / Offensive Language in Social Media

Abstract

The identification of offensive language on social media platforms has been a subject of ongoing research in recent years. In countries where English is not the native language, social media users often employ a code-mixed style in their posts and comments on YouTube, Twitter, Facebook, etc. This introduces numerous challenges for tasks related to identifying offensive content, particularly in languages with low resources such as Tamil, Urdu, Malayalam, etc. In this study, we conduct a comparative analysis of the performance of various machine learning and natural language processing models to detect offensive content on social media platforms. In our analysis, language category, social media platform from which they got the dataset, methodology, models used, and finally identified the outperformed model and benefits of the research were tabulated. Finally, the best model was identified among the analyzed models.