Feature Selection Techniques in Intrusion Detection: A Comprehensive Review


  • Lubna ALkahla Ninevah University
  • Maher Khalaf Hussein University of Telafer
  • Asmaa Alqassab University of Mosul




Intrusion detection, IDS, Feature selection, Feature Reduction, Cybersecurity, Security


This investigation aims to explore previous research on the implementation of feature selection in intrusion detection. Feature selection has demonstrated its ability to enhance or sustain comparable classification accuracy levels for intrusion detection systems, while simultaneously improving classification efficiency. The evaluation includes an assessment of filter-based, wrapper-based, and hybrid feature selection techniques. Given that Big Data challenges can affect intrusion detection, feature selection’s classification efficiency can aid in lowering computing requirements. Older KDD intrusion detection datasets have received considerable attention in previous feature selection research. Consequently, researchers need more high-quality datasets that are available to the general public.


Download data is not yet available.

Author Biographies

Lubna ALkahla, Ninevah University

Software Department, Information Technology College

Maher Khalaf Hussein, University of Telafer

Computer Center, Presidency of the University

Asmaa Alqassab, University of Mosul

Department of Computer Science, College of Education for Pure Science