Paper Title
A Hybrid Intrusion Detection System Using Particle Swarm Optimization For Feature Selection

The ultimate goal of this paper is to develop systems for intrusion detection in computer networks to achieve the best accuracy performance. This study suggests the idea of classifier combination. This hybrid approach is based on optimization and feature selection using a combination of two-step approach for the classification. In the first step, using Accelerated Particle Swarm Optimization (APSO) algorithm, a set of best discriminating features is selected. Then using a combination of three classifiers namely KNN, Decision Tree and Neural Network, intermediate data is generated. Finally these data is fed to the AdaboostM2 classifier for final decision. The performance of the proposed approach is evaluated with criteria such as F measure, Accuracy and False Alarm. Experiments on KDD-CUP99 dataset show the effectiveness of the proposed approach compared to the most recent approaches in this context. Index Terms- Intrusion Detection, Hybrid Classifier, Particle Swarm Optimization, Adaboost, Feature selection