DC-Tree: Density-Based Clustering Index for Objects in Skewed Distribution

Jine Tang,Dandan Li,Zhangbing Zhou,Lei Shu,Daqiang Zhang,Qun Wang

DOI: https://doi.org/10.1109/wetice.2012.27

2012-06-01

Abstract:Efficient spatial index is essential for querying data objects in spatial databases. Data objects may be unevenly distributed in real situations. In this setting, R-tree and its variants may cause large overlap and coverage among branch nodes, which impact the query efficiency to some extent. To address this challenge, this paper proposes a novel Density-based Clustering tree (DC-tree) by clustering data objects. Data objects in a dense region will be put into the same node. Thus, overlap and coverage among node regions are less than that of R-tree and its variants. Since dense regions contain more data objects, we assign a higher priority to these region nodes for facilitating the query operation. Experimental results show that in the context of skewed distribution, DC-tree can have a better performance for the insertion, deletion and query operations than that of traditional R-tree.