CyberSaver – A Machine Learning Approach to Detection of Cyber Bullying

Abstract

In this modern era of extensive use of online resources there has been reports of numerous cases of cyberbullying. Although awareness through medical health support systems such as counselling and psychological assistance is available, a system to combat threats is needed to handle the increasing rate of cyber bullying. This paper presents a model that can be used to detect and report cyberbullying with the use of machine learning techniques. A careful selection of the machine learning algorithms has been identified that could enable better accurate detection. The model was transformed into a prototype in python to evaluate the effectiveness of the model in detecting cyber bullying. The proposed model primarily focusses on test based and image-based threats as they are more common than other forms of cyber bullying.