A Deep Learning-Based Framework for Feature Extraction and Classification of Intrusion Detection in Networks

Table 1

Advantages and disadvantages of feature selection techniques.

Selection techniques	Advantages	Disadvantages	Methods
Filter	(i) Computationally efficient; (ii) Higher processing power; (iii) Independent of classification algorithm	Can select irrelevant features	(i) ANOVA; (ii) Chi square; (iii) Pearson correlation
Wrapper	(i) Highly accurate feature subset	High computational cost; requiring leaning algorithm	(i) Recursive technique; (ii) Forward selection; (iii) Backward elimination