Arabic cyberbullying detecting using ensemble deep learning technique

<span lang="EN-US">There has been a huge growth in recent years interest studies on abusive language and cyberbullying detection due to its effects both individual victims societies. Hate speech, bullying, racism, aggressive content, harassment other forms of abuse have all significantly increased as result Facebook, Instagram, social media platforms (SMPs). Since there is significant need detect, control, prohibit the circulation offensive content networking sites, we undertook this study automate identification or cyberbullying. Arabic data set balanced will be used process. Recently, ensemble machine learning increase effectiveness categorization models. more precise given that each spatial feature text can make references every contextual piece information. The authors utilized model merged convolutional neural network (CNN) with bidirectional long short-term memory (Bi-LSTM) inverse document frequency gated recurrent unit (GRU) hybrid fashion without any post-processing. Our work outperformed publicly released cutting-edge specifications official deep challenge. findings indicate three-layer (LSTM) classifier surpassed classifiers accuracy score 92.75% compared different algorithms.</span>