DefinePK

Abstract

The number of attacks in recent times has tremendously increased due to the increase in Internet activities. This security issue has made the Intrusion Detection Systems (IDS) a major channel for information security. The IDS’s are developed to in the handling of attacks in computer systems by creating a database of the normal and abnormal behaviors for the detection of deviations from the normal during active intrusions. The issue of classification time is greatly reduced in the IDS through feature selection. This paper is proposing the implementation of IDS for the effective detection of attacks. Based on this, the Firefly Algorithm (FA), a new binary feature selection algorithm was proposed and implemented. The FA selects the optimal number of features from NSL dataset. Additionally, the FA was applied with multi-objectives depending on the classification accuracy and the number of features at the same time. This is an efficient system for the detection of attacks reduction of false alarms. The performance of the IDS in the detection of attacks was enhanced by the proposed classification and feature selection algorithms.