摘要
在PSO聚类算法的基础上,提出了基于量子行为的微粒群优化算法(QPSO)的数据聚类。QPSO算法不仅参数个数少、随机性强,并且能覆盖所有解空间,保证算法的全局收敛。PSO与QPSO算法的不同在于聚类中心的进化上,实验中用到四个数据集比较的结果,证明了QPSO优于PSO聚类方法。在聚类过程中使用了一种新的度量代替Euclidean标准,实验证明了新的度量方法比Euclidean标准更具有健壮性,聚类的结果更精确。
A data clustering using quantum-behaved particle swarm optimization （QPSO） based on PSO clustering was proposed. Not only parameters of QPSO is few and randomicity of QPSO is strong, but also QPSO cover with all solution space and guarantees global convergence of algorithms. The difference between PSO and QPSO is the evolution of the cluster centroids. The performance of the clustering method on four data sets were compared. The experiment results show QPSO clustering superiority. A new metric was used to replace the Euclidean norm in clustering procedures. Experiment results show that this new metric is more robust and accuracy than common-used Euclidean norm.
出处
《计算机应用研究》
CSCD
北大核心
2007年第11期49-51,共3页
Application Research of Computers
基金
国家自然科学基金资助项目(60474030)
