Classification of Cyberbullying Text in Arabic
Benaissa Azzeddine Rachid, Azza Harbaoui, Hajjami Henda Ben Ghezala
Abstract
The increase in electronic devices and social media use has allowed face-to-face bullying integrate the cyber space. Cyberbullying is an increasing problem that affects its victims worldwide both mentally and physically. Acting upon this phenomenon is of highly importance. Several researches were conducted on cyberbullying classification in English language and less on Arabic. In this paper, we conducted a series of experiments using neural network models (Convolutional and Recurrent Neural Networks) and pre-trained word embeddings in an attempt to classify cyberbullying instances on an Arabic channel news comments dataset. Best models achieved 0.84 F1-score on a balanced version of the aforementioned dataset.