Intrusion Detection Using Deep Learning

Intrusion detection is deemed to be a cornerstone of cyber security. Early and effective intrusion detection has been attracted much attention from researchers in the last decade. However, the existence of a deep and adequate study in using deep learning models for intrusion detection in cyber security is still seldom. In this study, I have investigated the problem of intrusion detection in three different environments, namely, personal computer, network and cloud computing. Furthermore, a double Particle Swarm Optimization-based algorithm is proposed for both feature and hyperparameter selection. Finally, a novel deep learning approach is presented to improve the performance of intrusion detection in cyber security area.