模糊c均值聚类算法是目前聚类分析中最受欢迎的算法之一,但其聚类效果往往受初始参数的影响。针对这一问题,提出一种基于网格和密度的模糊c均值聚类初始化方法。以网格和密度为工具提取聚类样本的类聚类中心,以此来初始化模糊c均值聚类算法的初始参数,从而弥补原算法的不足。实验证明方法是可行的、有效的;
Fuzzy c-means clustering algorithm is one of the most widespread clustering algorithm, Its performance strongly depends on the initial parameters. To solve this problem, an initialization method for fuzzy c-means clustering algorithm based on grid and density is proposed, Grid and density are used to extract the clustering centers of samples, and initialize the initial parameters of fuzzy c-means clustering algorithm. Experiment shows that this method is feasible and valid.