Spatial Clustering Algorithm Based on Hierarchical-Partition Tree

In spatial clustering, the scale of spatial data is usually very large. Spatial clustering algorithms need high performance, good scalability, and are able to deal with noise and multidimensional data. In this paper, we propose a rapid spatial clustering algorithm based on hierarchical-partition tree. The proposed algorithm partitions spatial data into subsets by simple arithmetical calculation and set calculation, which are separately based on single-dimensional distance and set-indices. At the same time, we propose a novel spatial indexing technology named hierarchical-partition tree to store and search spatial data. Our experimental results on both synthetic and real-world data show that the new algorithm not only has a very high efficiency, but also can deal with clusters of any shaped and highdimensional data. And it is not sensitive to noise data.