Effective Intrusion Detection Systems using Genetic Algorithm

Abstract

Conventional methods of intrusion prevention like firewalls, cryptography techniques, have not proved themselves to completely defend networks and systems from newly generated malwares and attacks. Intrusion Detection Systems (IDS) are useful to find the correct solution to solve the current problems and became an important part of any security network infrastructure to detect these threats without generating any problem to network. The basic purpose of IDS is to detect attacks and their nature that may harm the computer system. Several different approaches for intrusion detection are available as per the literature. These approaches are broadly defined by three ways: i) Signature based approach ii) Anomaly based approach and iii) Hybrid approach that combines signature and anomaly detection approaches. The proposed system works for signature based concept using genetic algorithm as features selection and detection .The system is tested on KDDCup99 and NSL-KDD dataset using Weka3.6 classifiers and implemented classifier.