Automated Cyberstalking Classification using Social Media

Cyber Bullying (CB) is now a deliberate and forceful action by one and more individuals or groups against an individual or group through electronic media. Automated detection tools are available, but quality datasets or insufficient tools are lacking. The various features are spread during CB detection. An integrated model that integrates both feature extraction is presented in this paper. The data pipeline incorporates raw text datasets from social media, a motor and a classification motor. Features of extracts from the user's features, remarks, and context sbeconsiderederedt when determining whether CB is present. An engine of classification based on the resiss are claisfied as the Artificial Neural Network (ANN) and an ,assessment system that pays or penalizpenalisesvided a classwithified information document, ; deepnhancement learning (DRL) is used to improve system performance during the evaluation of a system, the ANN-DRL is validated with various metrics, including classification (consisting of) Precision, reminder, and f-measurement. The simulation results demonstrate that the ANN-DRL is effective and works better and conventional classifiers for machine learning can be more precise.