Abstract:
With the deficiency of global search ability for K-means clustering algorithm, K-means algorithm based on particle swarm optimization (PSO-KM) is proposed. As an evolutionary computation technique based on swarm intelligence particle swarm optimization (PSO) algorithm has high global search ability, the analysis and experiment show PSO-KM could avoid local optima and has relatively good global convergence. Experiment over network connection that records from KDD CUP 1999 data set is implemented to evaluate the proposed method. The results clearly show the outstanding performance of the proposed method.
Key words:
Particle swarm optimization (PSO),
K-means algorithm,
Global optimization,
Intrusion detection
摘要: 针对K均值聚类算法在全局优化中的不足,提出了基于粒子群的K均值(PSO-KM)聚类算法。粒子群优化算法作为一种基于群智能方法的演化计算技术,有很好的全局搜索能力。通过理论分析及实验证明,该算法有较好的全局收敛性,能有效地克服传统的K均值算法易陷入局部极小值的缺点。对KDD-99数据集的仿真实验结果表明,该算法在入侵检测中能获得令人满意的检测率和误检率。
关键词:
粒子群优化,
K均值算法,
全局优化,
入侵检测
XIAO Lizhong; SHAO Zhiqing; QIAN Xiyuan. A Hybrid Clustering Algorithm for Network Intrusion Detection[J]. Computer Engineering, 2007, 33(04): 125-127.
肖立中;邵志清;钱夕元. 一种用于网络入侵检测的杂交聚类算法研究[J]. 计算机工程, 2007, 33(04): 125-127.