Introduction: The distributed structure of cloud computing makes it an attractive target for potential cyberattacks by intruders. In this paper, using the anomaly detection approach, a method for embedding an intrusion detection system for cloud computing is presented.
More
Introduction: The distributed structure of cloud computing makes it an attractive target for potential cyberattacks by intruders. In this paper, using the anomaly detection approach, a method for embedding an intrusion detection system for cloud computing is presented. Therefore, by studying how to check the parameters and the combined role of the parameters in the detection of penetration in the cloud, a method for detecting suspicious behavior in the cloud is provided. The most logical way to detect an intrusion is to use supervised methods to learn the parameters of normal customer behavior. Therefore, the detection of biased behavior in the form of suspicious behavior was implemented and discussed, investigated, and compared with an initial simulation in the form of identifying abnormal behavior in different behavioral areas by the neural network.Method: In this article, the basis of abnormality detection in different aspects is to examine the behavior of users and use the capabilities of reproducing inputs in RNN neural networks. In these networks, during the training of the network, the weights are adjusted in such a way that they can minimize the average square of the error so that the network can produce common repeating patterns well. Therefore, after training, these networks cannot reproduce well the input patterns that are actually significantly different from the training samples. Hence, these networks are able to identify anomalies in the tested sets. Accordingly, RNN networks are used here to model normal behavior.Findings: The simulation results show that the proposed method, which is based on the recurrent neural network, can improve the false positive, false negative, and detection accuracy compared to the classification method.Discussion: In this article, the detection of biased behavior in the form of suspicious behavior was implemented and discussed, investigated, and compared with an initial simulation in the form of identifying abnormal behavior in different behavioral fields. The simulation results show that the proposed method, which is based on the iterative neural network, can improve the false positive, false negative, and detection accuracy compared to the classification method.
Manuscript profile