Skip to main content

Table 2 Evolving Attacks in the UNSW_NB15 dataset

From: Conceptualisation of Cyberattack prediction with deep learning

Attack ClassDescriptionNumber of Instances
NormalBenign traffic56,00037,000
AnalysisCan be a port scan or spam2000677
BackdoorBypassing a secured authentication to have access to a machine1746583
DoSSee TableĀ 112,2644089
ExploitsExploitation of a vulnerability in a piece of software33,39311,132
FuzzersFeeding a network with randomly generated data to cause it to malfunction18,1846062
GenericAn attack on block ciphers40,00018,871
ReconnaissanceAn attack that extracts information about a user or network10,4913496
ShellcodeExploitation of a software vulnerability using a small piece of code as payload1133378
WormsAn attack that can replicate itself across multiple connected systems or networks13044