面向非球形分布数据的自适应K近邻聚类算法

Adaptive K Near Neighbor Clustering Algorithm for Data with Non-spherical-shape Distribution

针对传统聚类算法处理非球形分布数据的不足，提出了一种新型的自适应K近邻 聚类算法。该算法由数据集归一化、初始类别构造和初始类别融合3个步骤构成。仿真结果 表明，该算法在无须聚类数目的前提下，对非球型分布数据具有很好的聚类效果。

To the shortage of traditional clustering algorithm when dealing dat a with non-spherical-shape distribution, a novel adaptive K near neighbor cluste ring algorithm is presented in this paper. This algorithm is made up of three pa rts: (a)uniform for data; (b) constitution of initial patterns; (c)fusion of in itial patterns. The simulation results show that this algorithm has good cluster ing performance for data with non-spherical-shape distribution without knowing t he number of clustering.