Cyber Bullying Detection Using Machine Learning

Abstract

Cyber bullying has evolved as a severe problem hurting children, teenagers, and young adults as a result of the increasing use of social media. Automatic detection of bullying communications in social media is now possible, thanks to machine learning techniques, which could aid in the creation of a healthy and safe social media environment. One major issue in this important research area is robust and discriminative numerical representation learning of text messages. To address this challenge, we offer a new representation learning method in this study. The Semantic-Enhanced Marginalized Denoising Auto-Encoder (SMSDA) is a semantic enhancement of the popular deep learning model stacked denoising Auto-Encoder. The semantic extension is made up of semantic dropout noise and sparsity constraints, with the semantic dropout noise being the most important.