A Novel Framework Design of Network Intrusion Detection Based on Machine Learning Techniques

Table 8

Comparison of the proposed framework in the binary classification with related work using CICIDS2017 dataset evaluation and some of the traditional ML approaches mentioned in their papers.

	Reference + method	Accuracy	Precision	Recall	F1-score
[8]	MDAE + LSTM	0.9990	0.9990	0.9980	0.9990
	Naive Bayes	0.3130	0.3000	0.9790	0.4590
	SVM	0.7990	0.9920	0.3280	0.4930
	DNN	0.9310	0.8270	0.9740	0.9840
	MDAE	0.9140	0.8980	0.8020	0.8400
	LSTM	0.9980	0.9990	0.9970	0.9980
[24]	DNN + without IPs			0.9677
[18]	Feature selection + ID3	0.9515	0.9500	0.9500	0.9500
[23]	PCA + RF	0.9960		0.9880	0.9970
	AE + RF	0.9950		0.9850	0.9960
	Our approach	0.9992	0.9992	0.9992	0.9992

Note. The best values are in bold, and missing value means not provided in reference.