Hate Speech Classification Using Rfdt, Bilstm, and Bilstm

Abstract

In this article, we will consider how to classify hate speech on social media. To avoid conflicts between citizens, it is necessary to automatically detect the spread of hate speech and offensive words on social media. Additionally, hate speech has targets, categories, and dimensions that need to be recognized in order for government agencies to prioritize which hate speech needs immediate action. This article describes the classification of hate speech and offensive words in texts on the social media: Twitter (X platform), English, and a mixture of both, down to types, categories, and levels. Classification of hate speech multi-label texts is studied using his RFDT and BiLSTM with pre-trained BERT models.