A Survey on Feature Selection for Intrusion Detection  
Author Richard Zuech


Co-Author(s) Taghi M. Khoshgoftaar


Abstract This study examines previous research works in applying feature selection to Intrusion Detection. Feature selection has proven to improve or maintain similar classification accuracy for Intrusion Detection Systems (IDSs) while improving classification efficiency. Wrapper-based, Filter-based, and Hybrid feature selection techniques are evaluated. Since Intrusion Detection can face Big Data challenges, the classification efficiencies provided by feature selection can reduce computational demands. Previous feature selection research has been too narrowly focused on older KDD Intrusion Detection data sets. Researchers need access to additional high quality data sets which are publicly available.


Keywords Intrusion detection, IDS, Feature selection, Feature Reduction, Cybersecurity, Security
    Article #:  21150
Proceedings of the 21st ISSAT International Conference on Reliability and Quality in Design
August 6-8, 2015 - Philadelphia, Pennsylvia, U.S.A.