Extremely boosted neural network for more accurate multi-stage Cyber attack prediction in cloud computing environment

• Published in: Journal of cloud computing : advances, systems and applications Vol. 12; no. 1; pp. 14 - 22
• Main Authors: Dalal, Surjeet, Manoharan, Poongodi, Lilhore, Umesh Kumar, Seth, Bijeta, Mohammed alsekait, Deema, Simaiya, Sarita, Hamdi, Mounir, Raahemifar, Kaamran
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.12.2023; Springer Nature B.V; SpringerOpen
• Subjects: Accuracy; Algorithms; Bayesian analysis; Bayesian network; Cloud computing; Computer Communication Networks; Computer Science; Computer System Implementation; Computer Systems Organization and Communication Networks; Cybersecurity; Information Systems Applications (incl.Internet); Intrusion detection; Machine learning; Multi-stage cyber attack; Neural network; Neural networks; Quest; Recent Advancements in Cloud Security using Performance Technologies and Techniques; Software Engineering/Programming and Operating Systems; Special Purpose and Application-Based Systems; Zero-day attack; Multi-stage cyber attack; Intrusion detection; Zero-day attack; Neural network; Bayesian network; Quest; Security Investigation
• ISSN: 2192-113X; 2192-113X
• DOI: 10.1186/s13677-022-00356-9

There is an increase in cyberattacks directed at the network behind firewalls. An all-inclusive approach is proposed in this assessment to deal with the problem of identifying new, complicated threats and the appropriate countermeasures. In particular, zero-day attacks and multi-step assaults, which are made up of a number of different phases, some malicious and others benign, illustrate this problem well. In this paper, we propose a highly Boosted Neural Network to detect the multi-stageattack scenario. This paper demonstrated the results of executing various machine learning algorithms and proposed an enormously boosted neural network. The accuracy level achieved in the prediction of multi-stage cyber attacks is 94.09% (Quest Model), 97.29% (Bayesian Network), and 99.09% (Neural Network). The evaluation results of the Multi-Step Cyber-Attack Dataset (MSCAD) show that the proposed Extremely Boosted Neural Network can predict the multi-stage cyber attack with 99.72% accuracy. Such accurate prediction plays a vital role in managing cyber attacks in real-time communication.