This paper investigates the performance of different feature selection techniques such as ranking and subset-based techniques, aiming to find the optimum collection of features to detect attacks with an appropriate… Click to show full abstract
This paper investigates the performance of different feature selection techniques such as ranking and subset-based techniques, aiming to find the optimum collection of features to detect attacks with an appropriate classifier. The results reveal that more accuracy of detection and less false alarms are obtained after eliminating the redundant features and determining the most useful set of features, which increases the intrusion detection system (IDS) performance.
               
Click one of the above tabs to view related content.