A Feature Selection Technique for Cloud IDS Using Ant Colony Optimization and Decision Tree

Cloud computing is a new computing paradigm offering computing resources as a service over the internet. It is a promising and emerging IT technology with enormous potentials and benefits to customers; however there are underlying security issues and vulnerabilities militating against its wide acceptance. Hence, providing effective security is paramount to cloud computing due to security threats arising from resource outsourcing. Intrusion detection is a vital security measures used to ensure cloud security; however, the performance of IDS is hampered by irrelevant features in the network traffic packet; hence, feature selection is employed to select the minimum subset of features capable of representing the original features and also provide high classification accuracy. However, the existing feature selection technique are inadequate in finding the optimal subset. Therefore, in this paper a new feature selection technique is proposed that employs Ant Colony Optimization and Decision Tree to enhance detection accuracy of cloud IDS. Experimental results conducted using dataset generated from ESX hypervisor and vSphere show that the proposed feature selection technique outperforms existing feature selection techniques proposed for Cloud IDS such as Genetic Algorithm and Rough Set.