大型数据库中基于邻域连接的层次聚类算法

提出了一种基于邻域连接的层次聚类算法HANL,该算法首先采用分割的方法将数据集划分为若干个子簇,通过对子簇间的连接的分析,建立子簇间的连接构成图,图中带权的边代表了子簇间的连接紧密度。合并连接紧密度高的结点,得到最后的聚类结果。该算法适用于高维数据集,能够对任意形状的簇进行聚类,无论对于数值属性的数据库还是分类属性的数据库都是一个有效的聚类方法。同时这种方法聚类速度快,效率高,具有良好的伸缩性。

Hierarchical Algorithm based on Neighborhood-Linked in Large Spatial Databases

This paper presents a novel hierarchical clustering algorithm based on neighbor linked which uses partition-ing method to divide the data set into several sub_clusters.By analyzing the link of the sub_clusters and transfering it to the graph model where the edges with weight represent the joint degree of sub_clusters.Combining the nodes which have the high joint degree,we can get the result of the clustering.The algorithm can be performed in high-dimensional data set,clustering the arbitrary shape of clusters such as the spherical,linear and elongated etc.Because of the fast running speed,high efficiency,this algorithm is available for large spatial databases.