Empirical Evaluation of Noise Influence on Supervised Machine Learning Algorithms Using Intrusion Detection Datasets

Table 17

Using ensemble learning (bagging and boosting).

Ensemble learning	ML algorithm	NSL-KDD			UNSW-NB15
		Accuracy	Precision	Recall	Accuracy	Precision	Recall
Baseline	DT (J48)	81.5339	0.858	0.815	100	1.000	1.000
	RF	80.4516	0.852	0.805	98.4903	0.985	0.985
	SVM	75.3948	0.802	0.754	100	1.000	1.000
	ANN	77.7147	0.817	0.777	100	1.000	1.000
	NB	76.1178	0.809	0.761	87.4435	0.884	0.874
Bagging	DT (J48)	83.7252	0.870	0.837	100	1.000	1.000
	RF	80.1499	0.850	0.801	99.9976	1.000	1.000
	SVM	75.0311	0.800	0.750	100	1.000	1.000
	ANN	76.0956	0.808	0.761	100	1.000	1.000
	NB	76.2952	0.810	0.763	87.2929	0.882	0.873
Boosting	DT (J48)	77.8522	0.838	0.779	100	1.000	1.000
	RF	79.4092	0.846	0.794	90.5577	0.922	0.906
	SVM	75.6343	0.803	0.756	100	1.000	1.000
	ANN	77.7147	0.817	0.777	100	1.000	1.000
	NB	73.7447	0.799	0.737	99.949	0.999	0.999