Oppositeness-based Hate Speech Detection

With the huge amount of web data available in social media platforms, detecting hate speech (HS) on social media became critical. Even though many HS detection machine learning algorithms have been developed using modern concepts such as deep learning and artificial intelligence, HS detection continues to be a challenge because humans work out ways to fool the algorithms. The current approach is about a comparative analysis of Natural Language Processing (NLP) Model implemented in Transformer library such as bart-large-mnli by Facebook, xlm-roberta-large-xnli which is finetuned over several Natural language inference datasets, ready to use for zero-shot classification & etc.