Subspace clustering via graph auto-encoder network for unknown encrypted traffic recognition

Cybersecurity

Table 5 The results of SCGAE and four incomplete clustering models

Dataset	Methods	ACC	NMI	ARI	F1
Ours	SCGAE without graph	0.8345	0.7811	0.8190	0.6526
	SCGAE without content	0.8523	0.8012	0.8812	0.6725
	SCGAE without self-expression	0.7812	0.6254	0.7812	0.6712
	SCGAE without self-supervised	0.8123	0.7892	0.7738	0.6891
	SCGAE	0.9067	0.8679	0.9235	0.6922
ISCXVPN2016	SCGAE without graph	0.7341	0.7254	0.7254	0.5623
	SCGAE without content	0.7212	0.7217	0.7116	0.5123
	SCGAE without self-expression	0.7125	0.7623	0.7123	0.5451
	SCGAE without self-supervised	0.7274	0.7431	0.7368	0.6535
	SCGAE	0.8122	0.8124	0.7761	0.7042