Research of Network Intrusion Detection Based on Particle Swarm Optimization and Support Vector Data Description

Concerning the data set of high dimensions in intrusion detection, the new algorithm based on support vector data description (SVDD) which optimized by particle swarm optimization (PSO) was proposed. In the improved algorithm, at firstly the particle swarm optimization was used to remove redundant features and reduce the data dimension in support vector data description. And then, the support vector data description built a super sphere model to detect the attacks from internet by analyzing the network connection data. Results from the experiments with the KDD CUP'99 network data indicate that the method of PSO-SVDD is better than traditional one class classifiers algorithm, which can improve the efficiency of intrusion detection and reduce the false detection rate.