Efficiently mining co-location rules on interval data

Abstract

Spatial co-location rules represent subsets of spatial features whose instances are frequently located together. This paper studies co-location rule mining on interval data and achieves the following goals: 1) defining the semantic proximity between instances, getting fuzzy equivalent classes of instances and grouping instances in a fuzzy equivalent class into a semantic proximity neighborhood, so that the proximity neighborhood on interval data can be rapidly computed and adjusted; 2) defining new related concepts with co-location rules based on the semantic proximity neighborhood; 3) designing an algorithm to mine the above co-location rules efficiently; 4) verifying the efficiency of the method by experiments on synthetic datasets and the plant dataset of "Three Parallel Rivers of Yunnan Protected Areas". © 2010 Springer-Verlag.