Evaluate the Detection Model by leveraging the Real Time Dataset from Private Cloud Environment

Abstract

Cloud security has risen to prominence as a major concern in recent years. There are numerous security frameworks in use today to protect the environment.   Attackers, on the other hand, have always found a way into the organization. In addition to the existing frameworks, the security frameworks must be extended to include the new dataset. Publicly available data sets are no longer effective in detecting new types of assaults and are therefore useless. The UNSW-NB15 data set is used as an offline dataset in this research to train an algorithm to detect malicious network traffic. Furthermore, this work produces its own real-time data collecting by establishing a private cloud environment at the home lab, which serves as a functioning example of the suggested intrusion detection methods. We used VMware vSphere to generate logs from a real-time environment, as well as malicious and non-malicious traffic on virtual machines in a private cloud environment. This dataset is used as a test data set to assess the performance of the proposed model. The recommended model with UNSW-NB15 and real-time data set has a higher accuracy rate than other existing techniques.