Improving the Accuracy of Network Intrusion Detection with Causal Machine Learning

Table 19

Performance of different classifiers in UNSW-NB15 dataset under different types of cyberattacks.

Algorithm	Detection accuracy under different types of cyberattacks (1, 9)
	1	9
RS-KNN-CFS	0.9283	0.7869
TPE-KNN-CFS	0.9168	0.7654
RS-KNN-IGBS	0.9501	0.7869
TPE-KNN-IGBS	0.9450	0.7073
RS-RF-CFS	0.9209	0.7806
TPE-RF-CFS	0.9274	0.7915
RS-RF-IGBS	0.9198	0.7253
TPE-RF-IGBS	0.9198	0.7367
BRS	0.8717	0.8082
CMLN	0.9926	0.9229