For computing services, cloud computing is represented as the internet-based method whereas cloud users use the resources. In the cloud service, intrusion or attacks is considered as one of the main problems in the cloud environment because it corrupts the performance. The service is affected because of the various attacks, and it provides deceptive information also maximizes the false rate. A novel Modified CSA-based Levenberg-Marquardt Neural Network (MCS-LM NN) is proposed in this paper to recognize the intrusion behavior. At first, the cloud network experiences producing the clusters by exploiting the WLI fuzzy clustering model. This model attains the diverse count of clusters in that the data objects are clustered together. Subsequently, the clustered data is subjected to the MCS-LM NN, which is the integration of the Levenberg-Marquardt method of NN and Modified CSA. The CSA is exploited to update the weight, as well as it also exploits to ascertain the optimal weight to identify the malicious activity via training procedure. Hence, the diverse clustered data is subjected to the developed optimization technique. After training the data, the data requires to be aggregated. Then that data is subjected to the MCS-LMNN model, whereas the intrusion behavior is recognized. At last, the experimentation outcomes of the developed technique and the performance analysis are carried out using the False Positive Rate (FPR), accuracy, and True Positive Rate (TPR). Therefore, the developed model obtains superior accuracy and it assures improved detection performance.
Buy article PDF