Feature Selection Based on Cross-Correlation for the Intrusion Detection System

Table 7

Comparison of the MIFS, CFA, and CCFS methods for DT, SVM, KNN, and NB classifiers on the CIC-IDS2017 dataset with 78 features.

CIC-IDS2017 (78 features)
Classification method	MIFS				CFS				CCFS
	DT	SVM	KNN	NB	DT	SVM	KNN	NB	DT	SVM	KNN	NB
Accuracy	95.24	94.43	94.82	94.01	96.87	96.02	96.54	94.59	97.91	97.12	97.24	96.32
Precision	95.53	95.12	95.76	94.61	97.84	96.60	97.68	95.06	99.42	98.23	98.75	96.62
Recall	95.53	94.82	95.76	94.61	97.84	97.28	98.02	95.61	99.07	97.94	98.75	96.61
F1-score	95.53	94.97	95.76	94.61	97.84	96.94	97.85	95.33	99.24	98.09	98.75	96.62