A Review of Machine Learning and AI-Based Approaches to Detecting Cyberbullying on Social Media

Abstract

With rapid rise in the use of social media, cyber-bullying has become a significant concern, affecting individuals of all ages and backgrounds. Finding cyberbullying on social media is essential to building a secure and welcoming online community. The various approaches and techniques used to identify cyberbullying on social media are examined in this research, ranging from traditional keyword-based strategies to more sophisticated machine learning and NLP-based algorithms. We also draw attention to the difficulties and restrictions posed by the use of current techniques, such as their poor accuracy, lack of standardization, and privacy issues. We also go over the significance of addressing social and cultural disparities as well as the necessity of ethical concerns when identifying cyberbullying. Our review provides a comprehensive understanding of the current state of cyberbullying detection on social media, along with recommendations for future research. Ultimately, combating cyberbullying and fostering a more supportive and positive online environment depend on the development of more precise and reliable detection techniques