Intrusion Detection in Industrial Internet of Things Network-Based on Deep Learning Model with Rule-Based Feature Selection

Table 3

Detection rates for NSL-KDD and UNSW-NB15 dataset classes.

Dataset	DoS	Normal	Backdoor	Worm	Shellcode	Probe	Exploits	U2R	Salicode	Analysis		Fuzzer	Generic	Reconnaissance
NSL-KDD	99.2%	99.7%	—	—	—	99.0%	—	75.5%	—	—	94.8%	—	—	—
UNSW-NB15	97.3%	99.6%	95.2%	81.7%	90.2%	—	98.0%	—	90.3%	92.3%		67.1%	99.8%	92.0%