Efficiently Mining High Utility Co-location Patterns from Spatial Data Sets with Instance-Specific Utilities

摘要

Traditional spatial co-location pattern mining attempts to find the subsets of spatial features whose instances are frequently located together in some regions. Most previous studies take the prevalence of co-locations as the interestingness measure. However, it is more meaningful to take the utility value of each instance into account in spatial co-location pattern mining in some cases. In this paper, we present a new interestingness measure for mining high utility co-location patterns from spatial data sets with instance-specific utilities. In the new interestingness measure, we take the intra-utility and inter-utility into consideration to capture the global influence of each feature in co-locations. We present a basic algorithm for mining high utility co-locations. In order to reduce high computational cost, some pruning strategies are given to improve the efficiency. The experiments on synthetic and real-world data sets show that the proposed method is effective and the pruning strategies are efficient.