Cloud-based intrusion detection using kernel fuzzy clustering and optimal type-2 fuzzy neural network

Nowadays, digital data is an important asset for every organization. With the advent of cloud computing, cloud service providers (CSPs) offer the required infrastructure to end-users for storage and provide flexibility in accessing data. Since the users access the data from the cloud through the Internet, the data stored in the cloud are exposed to various intrusions. Intrusion detection is considered to be a significant issue in the cloud. The existing techniques are capable to detect well-known attacks but fall short in detecting low frequent attacks. To address this issue, we propose a novel intrusion detection system (IDS) in the cloud using a combination of kernel fuzzy c-means clustering (KFCM) and an optimal type-2 fuzzy neural network (OT2FNN). To achieve this, we optimally select the parameters of T2FNN using the lion optimization algorithm (LOA) for weight optimization. The proposed IDS detects the intrusion and allow only normal data to be stored in the cloud. Simulation results on the NSL-KDD dataset show that the proposed IDS system gives better results than the existing IDS systems in terms of precision, recall, and F-measure.