Performance Investigation of Principal Component Analysis for Intrusion Detection System Using Different Support Vector Machine Kernels

The growing number of security threats has prompted the use a variety techniques. most common tools for identifying and tracking intruders across diverse network domains are intrusion detection systems. Machine Learning classifiers have begun to be used in threats, thus increasing systems’ performance. In this paper, investigation model an systems based on Principal Component Analysis feature selection technique different Support Vector kernels classifier is present. impact various kernel functions Machines, namely linear, polynomial, Gaussian radial basis function, Sigmoid, investigated. performance measured terms accuracy, True Positive, Negative, Precision, Sensitivity, F-measure choose appropriate function Machine. was examined evaluated using KDD Cup’99 UNSW-NB15 datasets. obtained results prove that superior sigmoid both Obtained and, CUP’99 were 99.11%, 98.97%, 99.03%. datasets 93.94%, 93.23%, 94.44%.