Review Article

A Systematic and Comprehensive Survey of Recent Advances in Intrusion Detection Systems Using Machine Learning: Deep Learning, Datasets, and Attack Taxonomy

Table 5

Detailed explanation of IDS datasets.

DatasetsNumber of recordsCreated byCategories of attacksYear of creation

KDD995 millionDapraDoS
U2R
Probe
R2L
1999
NSL-KDD5 millionUniversity of New Brunswick (UNB)DoS
U2R
Probe
R2L
2000
ISCXIDS20122.4 millionInformation Security Centre of Excellence (ISCX) (UNB)DoS, brute force, and distributed denial of service (DDoS)2012
CIDDS33 million flowsMarkus Ring et al.Brute force, DDoS, and port scans2017
UNSW-NB152.54 millionUniversity of New South Wales (UNSW), Canberra. Australian Centre for Cyber Security (ACCS)Fuzzers, shellcode, analysis, worms, backdoors, reconnaissance, dos, generic, and exploits2015
CICIDS20172.8 millionCanadian Institute of Cyber Security (CIC)Brute force FTP, Heartbleed, brute force SSH, DDoS, infiltration dos, web attack, and botnet2017
CSE–CIC–IDS201816.2 millionCommunications Security Establishment (CSE). Joined with ISCX and CICBrute force, DDoS, and web attacks2018
Ton_IoT22.3 millionSchool of Engineering and Information technology (SEIT), University of New South Wales (UNSW), CanberraDDoS, ransomware, data injection, DoS, password attack, cross-site scripting (XSS), backdoor, man-in-the-middle (MITM), and scanning2020