Spatial Co-location Pattern Mining Based on Density Peaks Clustering and Fuzzy Theory

Abstract

Spatial co-location patterns are the subsets of spatial features whose instances are frequently located together in geographic space. Traditional co-location pattern mining framework usually determines the proximity relationship of spatial instances by a user-specific distance threshold. However, in real life, the proximity relationship is a fuzzy concept and difficult to measure only by an absolute distance threshold. Furthermore, the spatial clique generating process consumes huge computational and spatial costs. In this paper, we propose a new framework for mining co-location patterns based on density peaks clustering and fuzzy theory. The experiments show that our method performs more efficient than the traditional Join-less method and the mining results on two real-world data sets indicate our method is significant and practical.