首页 > 代码库 > 笔记Clustering by fast search and find of density peaks
笔记Clustering by fast search and find of density peaks
We propose an approach based on the idea that cluster centers are characterized
by a higher density than their neighbors and by a relatively large distance from points with
higher densities. This idea forms the basis of a clustering procedure in which the number of
clusters arises intuitively, outliers are automatically spotted and excluded from the analysis, and
clusters are recognized regardless of their shape and of the dimensionality of the space in which
they are embedded.
想法来源是直接:1、cluster的中心密度要高;2、高密度的中心点之间的距离应该相对远一些。异常点都会被排除,同时也和形状无关。
问题来了,密度怎么定义?
dc是阶段距离,阈值啦。就是这个范围内有多少个点啦。后面就说这个值的选取鲁棒性不错。
简而言之,寻找比i节点密度大同时距离最近的点的距离为i的距离。
所以说如果密度大,同时又和其他密度比他大的点距离远,那么他很可能就是一个cluster的中心。
对于那个密度最大的点,定义 也就是离他最远的点的距离,默认他就是一个cluster的中心。
明天继续补充。
笔记Clustering by fast search and find of density peaks