Performance analysis of Word Embeddings for Cyberbullying Detection

Abstract

Cyber bullying activities are increasing day by day with the increase of Social Media Platforms such as Face book, Twitter, Instagram etc. Bullies take the advantage of these large online connected platforms due to which it became as a big challenging task in Natural Language Processing (NLP). In this paper, we compare the performance of various word embedding methods from basic word embedding methods to recent advanced language models such as RoBERTa, XLNET, ALBERT, etc. for cyberbullying detection. We used LightGBM and Logistic regression classifiers for the classification of bullying and non-bullying tweets. Among all the models, RoBERTa is outperformed as compared to state-of-the-art models.