A hybrid cloud-based intrusion detection and response system using gray wolf optimization (GWO) algorithm and artificial neural network (ANN)

Technology is growing rapidly and the use of cloud computing is increasing at the same speed. Most large and small companies use the cloud these days. Cloud computing provides economic benefits because it works with a pay-as-you-go logic. Security problems in the cloud have increased with the increase in cloud usage. To solve these problems, some mechanisms such as firewall, vulnerability scanners and intrusion detection system (IDS) and other methods to mitigate attacks are used. However, in case of new and previously unknown attacks, such systems are insufficient to detect attacks against the cloud. There are various security methods to protect cloud security from threats and vulnerabilities. In this thesis, a new hybrid cloud-based Intrusion Detection System (IDS) based on Gray Wolf optimization (GWO) and Artificial Neural Network (NN) has been proposed in order to detect attacks and ensure system security on the cloud. GWO is one of the meta-heuristic algorithms used effectively in many fields such as security, medicine / health, optimization, engineering and computing. In the thesis, GWO was used to train the Artificial Neural Network (NN) and the results were compared with other classification algorithms. In experimental studies, the current intrusion detection data sets such as UNSW-NB15 and NSL-KDD were used. The simulation results showed that the proposed algorithm increased the accuracy of intrusion detection by 97.83% and 98.21% for the NSL-KDD and UNSW-NB15 datasets, respectively.