中文说明:聚类分析的目的是将一组给定的数据分成子集,这样这些子集代表数据本身的某些相似之处。人类的眼睛立刻认识到两个几何形状、 两个半-月亮,并能够将数据划分为两个群集,在那里在同一点群集属于同一半月形。然而,一般来说,和特别是来自现实世界的问题的数据,是不可能只是看看数据,所以我们需要依靠算法来做到这一点。
English Description:
The objective of cluster analysis is to divide a given set of data into subsets,such that those subsets represent a certain similarity of the data itself.The human eye instantly recognizes two geometric shapes, two half-moons,and is able to divide the data into two clusters, where points within the samecluster belong to the same half-moon. However, in general, and especially with data from real world problems, it is not possible to just look at the data, so we need to rely on algorithms to do this.