基于免疫粒子群算法的特征选择

Feature selection based on immune particle swarm optimization

针对入侵检测中数据维数较高的问题,使用免疫粒子群算法Immune_PSO进行特征选择,消除冗余属性、降低问题规模、加快入侵检测速度。Immune_PSO算法使用二进制字符串序列来表示粒子位置,采用免疫算法思想进行粒子的选择,保持粒子的多样性,提高PSO算法的收敛精度。最后算法在KDD CUP1999数据集上进行了仿真实验,达到了预期的效果。

Concerning the data set with high dimensions in intrusion detection, an Immune_PSO algorithm based on immune and particle swarm optimization was proposed, which can select the most important features for intrusion detection, eliminate the redundancy property, reduce the problem size, improve the quality of classification and speed up the detection. The position of the particle was expressed with a binary string in Immune_PSO algorithm, and the selection of the particles was achieved by immune algorithm which can retain the diversity of particle and enhance the convergence results of PSO. The experiments with the KDD CUP1999 show that the proposed algorithm is efficient for feature selection.