Abstract
SummaryThe intrusion detection process is considered an efficient mechanism that plays a prominent role in network security. Traditional approaches for attack detection, like signature‐based detection, are easily prone to a new variant of attacks. Hence, the most prominent solution for cloud security attacks is an intrusion detection system through deep learning (DL) techniques. The behavior‐based intrusion detection system has been constructed by depicting the cloud traffic as input attributes. The deep learning technique is preferred to extract and analyze the traffic attributes rather than machine learning. Hence, an unsupervised technique named deep autoencoder (AE) is used to extract a feature in this research work. The resultant features are then classified by three classifiers named convolutional neural network (CNN), recurrent neural network (RNN), and long‐short term memory (LSTM). The designed approach is trained and assessed using a real‐time cloud attack dataset, namely, CICIDS 2018. Experimental results are analyzed using quality metrics such as receiver operating characteristics (ROC) curve, precision, accuracy, recall, and F1‐measure. Also, the performance of the proposed method is analyzed with existing literature. Higher accuracy rates like 99.55%, 94.33%, and 99.08% are attained with CNN, RNN, and LSTM networks, which are comparatively higher than the existing approaches.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: Concurrency and Computation: Practice and Experience
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.