Cyber Crime Prediction Using Machine Learning

With the plethora of information exchange in online communication, occurrences of some unwanted vices like cybercrimes and cyber bullying have taken shape. It is imperative to develop a prediction model for cyber-crime prediction. The existing technical approaches to identify these unwarranted activities have not achieved the required success rate. In the proposed approach, methods of Machine Learning and Neural Network are used for analyzing and predicting the occurrence of cybercrime. We have considered four major categories of cybercrime, namely cyber bullying, cyber harassment, cyber fraud and cyber hacking for performing various experiments. In the proposed model we have achieved an accuracy of 97.6% while predicting cyber-criminal activities.