Abstract: Spatial clustering has a significant application value in spatial data mining.In this paper,a novel clustering algorithm DBSB (Density Based Spatial Clustering Method with Heuristically Selecting Border Object) is proposed.The algorithm fastly expands the clusters by a heuristical function to choose core objects in the border region of the known core object,and then merge some clusters by border objects.That is,the DBSB algorithm get the ultimate clustering result through two steps of clustering.Finally,The theoretical analysis and experimental results indicate that the algorithm is effective and efficient.

摘要： 空间聚类在空间数据挖掘中具有重要的应用价值。本文提出了一种启发式选择边界对象的快速空间聚类算法DBSB,通过一个启发式函数近似选择相对于某个已知核心对象边界区域中的核心对象和边界对象,通过核心对象的序列来快速地扩展它们所在的簇,直至找到一些较小的簇。在此基础上再通过边界对象快速地合并某些簇,即该算法通过两步聚类,达到最终的聚类。理论分析和实验结果表明该算法有效可行。