摘要
从样本点的核密度估计出发,对集聚型点模式的集聚中心的个数和位置的确定方法进行了探索,提出一种集聚中心的核估计算法.与原有的基于几何概率提取集聚中心的方法相比,该算法对只有一个集聚中心的情况以及任意维数的样本空间点具有更好的估计效果.
We start with the kernel density estimation of data points to grope for the clustering center points, including computing the numbers and locations of the clustering center points, thus the kernel estimation method of clustering center points is put forward. Compared with the method based on geometrical probability, the arithmetic in this thesis is effective and better when there is only a clustering center point and when the data points have any dimension.
出处
《福州大学学报（自然科学版）》
CAS
CSCD
北大核心
2009年第3期322-325,共4页
Journal of Fuzhou University(Natural Science Edition)
关键词
集聚型
集聚中心
核密度估计
clustering type
clustering center point
kernel density estimation
