Deep learning for network intrusion: A hierarchical approach to reduce false alarms

• DOI: 10.1016/j.iswa.2023.200215
• Publication Date: 2023-03-15T17:49:25Z
• Authors: Samuel J. Moore; Federico Cruciani; Chris D. Nugent; Shuai Zhang; Ian Cleland; Sadiq Sani
• Keywords: Electronic computers. Computer science; Machine learning; 0202 electrical engineering, electronic engineering, information engineering; Q300-390; Deep learning; Network security; QA75.5-76.95; 02 engineering and technology; Network intrusion; Cybernetics

Deep learning Machine Network intrusion securityComputer networks form much of the infrastructure supporting day-to-day life in this digital age.Computer networks, however, are prone to attack and therefore require detection systems.Intrusion systems provide a mechanism detect network attacks at an early stage generate alerts.These systems, far from panacea.Rather, they tend overwhelm their operators with alerts, which more than 90% cases can be false positives.As such, problem positives is costly issue.This paper presents research design hierarchical detector, using deep learning, protects against raising vast numbers through implementation NIDS.This valuable advancement performance by reducing occurrence alarms 87.52%.The contained three contributions knowledge.The first these comparison between non-hierarchical understand would yield fewer alarms.The second contribution formulation approach, was able reduce 87.52%.Lastly, proposed model deployed live IoT environment, exposed genuine threats, environment analysed.