Finding N-most prevalent colocated event sets

Abstract

Recently, there has been considerable interest in mining spatial colocation patterns from large spatial datasets. Spatial colocations represent the subsets of spatial events whose instances are frequently located together in nearby geographic area. Most studies of spatial colocation mining require the specification of a minimum prevalent threshold to find the interesting patterns. However, it is difficult for users to provide appropriate thresholds without prior knowledge about the task-specific spatial data. We propose a different framework for spatial colocation pattern mining: finding N-most prevalent colocated event sets, where N is the desired number of event sets with the highest interest measure values per each pattern size. We developed an algorithm for mining N-most prevalent colocation patterns. Experimental results with real data show that our algorithmic design is computationally effective. © 2009 Springer Berlin Heidelberg.