Cyberbullying detection on social networking sites using BERT transformer.

Technological advancements have increased the amount of people using online social networking sites, which has led to an increase in cyberbullying. Online Social Networking sites provide a vast network for bullies to attack victims. Cyberbullying is a catch-all phrase for a wide range of online abuse, including but not limited to harassment, doxing and reputation assaults. These attacks often leave permanent mental scar(s) on the victim(s), which leads to drastic measures like depression, self-harm and suicidal thoughts. Given the consequences of cyberbullying, there is a dire need to take action against such crimes and to prevent them. This paper proposes a novel architecture to efficiently detect cyberbullying pattern. The proposed architecture utilizes the pre-trained model BERT to detect cyberbullying behavior on online platforms. The proposed models were tested on dataset taken from Kaggle and achieved accuracy of 80 percent. This paper provides a thorough examination of the various methodologies used for cyberbullying detection. [ABSTRACT FROM AUTHOR]