Cyber Bullying Detection on Social Media Network

Abstract

Abstract: Cyberbullying has grown more prevalent on social networking sites. Cyberbullying has resulted in a significant increase in mental health issues, particularly among the younger population. Self-esteem and mental health difficulties will affect a whole generation of young adults unless action is made to combat cyberbullying. Many classical machine learning methods have already been used for the automated identification of cyberbullying on social media. Since the advent of social media platforms about 20 years ago, there haven't been many effective ways to stop social bullying, and it has recently grown to be one of the most concerning problems. In this project, we create an AI-based system for detecting cyberbullying and investigate ways to identify fraudulent profiles and abusive speech on social media while separating them from ordinary vulgarity. Applying supervised classification techniques to a manually annotated open-source dataset, we seek to create lexical baselines for this job.