Unmasking Cyberbullying Through NLP-Driven Detection in Contemporary Social Media

This chapter delves into the critical issue of cyberbullying and the innovative use of Natural Language Processing (NLP) and Machine Learning to detect and mitigate this harmful behavior. It begins by highlighting the alarming rise of cyberbullying, its severe impacts on victims, and the various forms it can take, such as harassment, flaming, exclusion, and doxxing. The chapter then explores the role of NLP in understanding and analyzing digital communication, emphasizing its ability to uncover hidden meanings and sentiments in text. It also discusses the application of Machine Learning algorithms, particularly the Multinomial Naïve Bayes (MNB) theorem, in classifying and predicting instances of cyberbullying. The research methodology involves data validation, cleaning, vectorization using Tf-idf, and classification through MNB, achieving an accuracy of 75%. The chapter concludes by stressing the importance of combating cyberbullying to foster a safer and more respectful online community, and the potential of advanced technologies in this endeavor. Additionally, it provides insights into the communities most affected by cyberbullying, using word clouds and frequency distributions to highlight prevalent terms and issues. This chapter offers a comprehensive overview of the current state of cyberbullying detection and the promising role of NLP and Machine Learning in addressing this global challenge.