摘要
为了解决聚类算法容易陷入局部最优的问题,以及增强聚类算法的全局搜索能力,基于KHM算法以及改进的引力搜索算法,提出一种混合K-调和均值聚类算法(G-KHM)。G-KHM算法具有KHM算法收敛速度快的优点,但同时针对KHM算法容易陷入局部最优解的问题,在初始化后数据开始搜索聚类中心时采用了一种基于对象多样性及收敛性增强的引力搜索算法,该方法改进了引力搜索算法容易失去种群多样性的缺点,并同时具有引力搜索算法较强的全局搜索能力,可以使算法收敛到全局最优解。仿真结果表明,G-KHM算法能有效地避免陷入局部极值,具有较强的全局搜索能力以及稳定性,并且相比KHM算法、K-means聚类算法、C均值聚类算法以及粒子群算法,在分类精度和运行时间上表现出了更好的效果。
In order to solve the clustering algorithm is easy to fall into local optimum, and enhanced global search ability clus- tering algorithm based on gravitational KHM algorithm and improved search algorithm, this paper proposed a mixed K-harmo- nic means clustering algorithm. G-KHM KHM algorithm had the advantage of fast convergence, but for the KHM algorithm was easy to fall into local optimal solution, after initializing the data center at the start of the search cluster search algorithm used an object-gravity diversity and convergence enhancement. This method improved the shortcomings of gravitational search algo- rithm was easy to lose the diversity of the population, and also had a strong gravitational search algorithm global search capa- bility, allowed the algorithm to converge to the global optimal solution. Simulation results show that, G-KHM algorithm can ef- fectively avoid getting into local extremes, with a strong global search capability and stability, and compared to the KHM algo- rithm, K-means clustering algorithm, C-means clustering algorithm and particle swarm optimization in the classification accura- cy and run time shows a better effect.
出处
《计算机应用研究》
CSCD
北大核心
2016年第1期118-121,共4页
Application Research of Computers
基金
陕西省教育厅自然科学基金资助项目(12JK0557)
关键词
混合K-调和均值聚类
KHM算法
改进引力搜索算法
全局搜索能力
K-harmonic mixing means clustering
KHM algorithm
improved gravitational search algorithm
global searchcapability