LCHI: Low-Order Correlation and High-Order Interaction Integrated Model Oriented to Network Intrusion Detection

Table 4

The value distributions of the AWID, NSL-KDD, UNSW-NB15, CICIDS 2017, CICIDS 2018, and DAPT 2020 datasets.

Dataset	Attack type	Training set	Testing set
AWID	Normal	326311	104657
	Injection	13220	4294
	Impersonation	9853	3083
	Flooding	9731	3081
NSL-KDD	Normal	67343	9711
	DoS	45927	7458
	Probe	11656	2421
	R2L	995	2750
	U2R	52	200
UNSW-NB15	Normal	56000	37000
	Generic	40000	18871
	Exploit	33393	11132
	Fuzzers	18184	6062
	DoS	12264	4089
	Reconnaissance	10491	3496
	Analysis	2000	677
	Backdoor	1746	583
	Shellcode	1133	378
	Worm	130	44
CICIDS 2017	Benign	65093	32625
	DDoS	85404	42623
CICIDS 2018	Benign	445212	222414
	DoS-SlowHTTPTest	128779	64581
DAPT 2020	Benign	6003	2852
	Establish foothold	5621	2967
	Reconnaissance	34	10