Arabic cyberbullying detecting using ensemble deep learning technique

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