OpenCBD: A Network-Encrypted Unknown Traffic Identification Scheme Based on Open-Set Recognition

Table 9

When the number of encoders in the ensemble model of OpenCBD is 4, 8, and 12, accuracy, precision, recall, and F1-score of the 9-class classification tasks (identifying 8 known classes and 1 unknown class) vary with increasing percentage of known classes randomly selected.

	Accuracy			Precision			Recall			F1-score
4	8	12	4	8	12	4	8	12	4	8	12
60%	0.6746	0.6953	0.6976	0.7342	0.7544	0.7532	0.7104	0.728	0.7242	0.714	0.7363	0.7334
70%	0.6807	0.7207	0.7507	0.7523	0.7835	0.8031	0.6997	0.7326	0.7484	0.7159	0.7502	0.7685
80%	0.6846	0.7123	0.7253	0.7469	0.7779	0.7836	0.7124	0.7346	0.7357	0.7226	0.746	0.7533
90%	0.7192	0.7146	0.7361	0.771	0.7798	0.7939	0.7455	0.7317	0.7317	0.7519	0.7472	0.7526