Cellular Neural Network-Based Methods for Distributed Network Intrusion Detection

Table 3

The comparison of current machine learning based IDS policy in each detection node.

Title	Type	Characteristics	Limitations
	Statistics based	Adaptive blacklist-based packet filter [16]
	Data mining based	Frequency patterns (FP) and tree patterns (TP) [17]
Machine learning based IDS in each detection node		An additive decision rules binary classifier [18]	Processing time should be improved and the accuracy of intrusion detector should be maintained
	Classification based	Coarse-to-refined Grid Search Support Vector Machine (GS-SVM) [19]
		Lazy classifier by using genetic algorithm [20]
		Naïve Bayes and Neural network [21]