基于相对密度的多分辨率聚类算法

Relative Density Based Multi-resolution Clustering Algorithm

提出基于相对密度的多分辨率聚类算法,结合了密度聚类和模糊聚类的优点,能形成任意形状、多级分辨率的聚类结果,具有抗噪声能力和处理大数据集的能力,并有效地解决参数值难以设置,以及高密度簇完全被相连的低密度簇所包含等问题.

Provides a multi-resolution clustering algorithm based on relative density,which not only inherits the advantages of density based clustering and fuzzy clustering that can discover arbitrary-shape and multi-resolution clusters and deal with the large dataset,and are insensitive to noises,but also solves those common problems efficiently that clustering results are very sensitive to the user-defined parameters ,it is hard to determine reasonable parameters ,and high density clusters are contained fully in coterminous low density clusters.