Performance of Various SVM Kernels for Intrusion Detection of Cloud Environment
International Journal of Emerging Trends in Engineering Research · 2020
This paper investigates the performance among the various kernel based SVM classifiers for intrusion detection in cloud environment.Several researchers have presented the different kernel functions of SVM for Intrusion Detection.There is always an ambiguity in choosing which kernel function is to apply for better detection rate to identify classification accuracy factor.This paper explores to achieve this objective to identify the popular kernel functions linear, polynomial, radial basis function and Sigmoid.The CIDDS-001 dataset is adapted because of it is a recently available benchmark dataset and generated with new types of attacks of cloud environment.To evaluate the performance of different kernel functions computational time and accuracy taken as QoS metrics with ten-fold cross validation.The numerical results are calculated and conclusions are drawn.