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

Table 13

Results after conducting the noise filtering.

Noise filtering	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
Filtered	DT (J48)	80.7116	0.844	0.807	100	1.000	1.000
	RF	80.6742	0.844	0.807	99.9396	0.999	0.999
	SVM	80.6367	0.796	0.806	100	1.000	1.000
	ANN	80.6367	0.796	0.806	100	1.000	1.000
	NB	80.4869	0.746	0.805	75.4608	0.800	0.755