基于模拟退火的动态聚类算法

A Dynamic Clustering Algorithm Based on Simulated Annealing

提出了一个基于模拟退火的动态聚类算法,用模拟退火的全局优化能力来解决常规动态聚类方法中聚类结果对初始中心的敏感性问题。仿真结果表明,它是一个具有全局最优解的动态聚类方法。

A new dynamic clustering algorithm based on simulated annealing is presented. The problem of sensitivity with original clustering center is solved by the global optimization capability of the simulated an-nealing. The perfect performance is demonstrated by the simulation example.