Intrusion Detection Systems Based on Logarithmic Autoencoder and XGBoost

Table 5

Comparison of the proposed model and other excellent classifiers for UNSW-NB15.

Classifier	Accuracy	Precision	Recall	F1-score	Time (s)
MSCNN [21]	0.9140	-	-	-	571
MSCNN-LSTM [21]	0.9560	-	-	-	1,060
SaE-ELM-Ca [22]	0.8917	0.8079	0.9958	0.8920	2,168
SDAE-ELM3 [23]	0.7238	0.6994	0.8742	0.7771	5,943
XGBoost-DNN [8]	0.9950	0.9945	0.9942	0.9952	4,200
LogAE-XGBoost	0.9511	0.9549	0.9743	0.9645	132

It is used to distinguish the metrics of our proposed model from those of other models.