SUDBC:一种基于空间单元密度的快速聚类算法

随着数据规模越来越大,要求聚类算法有很高的执行效率,很好的扩展性,能发现任意形状的聚类以及对噪音数据的不敏感性.提出了一种基于空间单元密度的快速聚类算法SUDBC,该算法首先将被聚类的数据划分成若干个空间单元,然后基于空间单元密度将密度超过给定阈值的邻居单元合并为一个类.实验结果验证了SUDBC算法具有处理任意形状的数据和对噪音数据不敏感的特点.

SUDBC: a Quickly Clustering Algorithm Based on Spatial Unit Density

With the rapid increase of data scale in databases, it is required that clustering algorithms have high efficiency, extensibility, ability to deal with arbitrary shapes of clusters, and non-sensitivity to noise data. In this paper, we propose a quickly clustering algorithm SUDBC based on spatial unit density. It first partitions the data space to be clustered into a set of spatial units. Then, it merges all neighbor units into the same clusters based on the spatial unit density. Therefore, the algorithm can deal with the large scalability of points. Experimental results confirm that SUDBC algorithm has the feature of dealing with arbitrary shape of data and non-sensitivity to noise data.