Paper Title
Feature Extraction Using Sim-Swadorest Optimization Algorithm For Intrusion Detection

Abstract
Intrusion detection is a key element in system security that operates in real time network data. The Main drawback in detecting intrusion is its dataset size, which leads to a huge dimensional dataset problem. Therefore, feature extraction is necessary to reduce attribute size and improve detection rate to produce high efficiency. This paper proposed a new mechanism based on simplified swarm optimization (SSO) and Random Forest algorithm (RF) called Sim-Swadorest Optimization algorithm. SSO is used to find a more appropriate set of attributes for classifying network intrusions, and RF is used as a classifier to detect intrusion. The proposed method is investigated and assessed on live network data collected. It�s clear and consistently superior to other existing methods and improves the performance accuracy of filtering phase than PSO-RF and other classifiers. The test result shows that the proposed method reduces attribute size more effectively and achieves a near optimum solution. Index Terms�Swarm intelligence, Simplified Swarm Optimization, Particle Swarm Optimization, Random Forest, Intrusion detection.