Cybersecurity

Table 4 Selected features of UNSW-NB 15 and Network TON_IoT dataset

From: Performance analysis of machine learning models for intrusion detection system using Gini Impurity-based Weighted Random Forest (GIWRF) feature selection technique

UNSW-NB 15 (20 features)		Network TON_IoT (10 features)
feature name	Importance score	Feature name	Importance score
sttl	0.123	ts	0.258
ct_state_ttl	0.099	proto	0.146
sload	0.059	src_ip_bytes	0.137
rate	0.058	src_pkts	0.067
dload	0.057	dst_ip_bytes	0.062
dttl	0.045	dst_pkts	0.048
sbytes	0.039	conn_state	0.047
ct_srv_dst	0.037	dst_bytes	0.034
smean	0.034	src_bytes	0.032
dmean	0.032	duration	0.030
dbytes	0.028	–	–
ackdat	0.027	–	–
ct_dst_src_ltm	0.027	–	–
ct_srv_src	0.026	–	–
dur	0.026	–	–
tcprtt	0.025	–	–
synack	0.024	–	–
dinpkt	0.023	–	–
sinpkt	0.022	–	–
dpkts	0.021	–	–

Back to article page