Discovering colocation patterns from spatial data sets: a general approach

Abstract

Abstract-Given a collection of Boolean spatial features, the colocation pattern discovery process finds the subsets of features frequently located together. For example, the analysis of an ecology data set may reveal symbiotic species. The spatial colocation rule problem is different from the association rule problem since there is no natural notion of transactions in spatial data sets which are embedded in continuous geographic space. In this paper, we provide a transaction-free approach to mine colocation patterns by using the concept of proximity neighborhood. A new interest measure, a participation index, is also proposed for spatial colocation patterns. The participation index is used as the measure of prevalence of a colocation for two reasons. First, this measure is closely related to the cross-K function, which is often used as a statistical measure of interaction among pairs of spatial features. Second, it also possesses an antimonotone property which can be exploited for computational efficiency. Furthermore, we design an algorithm to discover colocation patterns. This algorithm includes a novel multiresolution pruning technique. Finally, experimental results are provided to show the strength of the algorithm and design decisions related to performance tuning.

Rule Discovery and Probabilistic Modeling for Onomastic Data

“…The interpretation of such a rule is that given a location (x, y) in which event of type A occurs, one is likely to see at least one event of type B within distance r from (x, y). This definition is close to the ones used by [5][6][7][8][9]. From an onomastic point of view it seems prudent to start with restricting ourselves to associations between two names.…”

“…Onomastically, this is rather surprising. It is true that our data set contains such names as Pahalampi "Evil Lake" 7 (shown in Figure 4) or Palolampi "Burnt Lake", 8 where there are no instances within 2 km of each other. However, the area covered by such selections is rather small, and most of these findings cannot be considered significant.…”

Discovery of Collocation Patterns: from Visual Words to Visual Phrases

“…There are related work in modeling textons [25] and learning generic parts from multiple object categories for object recognition [20,5,23]. In data mining domain, there are also related work in discovering spatial collocation patterns [9] and interpreting mined frequent patterns [24,14]. These methods are concerned on discrete data, and may not be directly applied to visual data.…”

On the Relationships between Clustering and Spatial Co-location Pattern Mining

“…Proximity between a pair of spatial features can also be defined by various spatially meaningful measures based on their objects including participation index [9], join selectivity, G nearest neighbor function, F nearest neighbor function, J function, Ripley's K function and its variations [3]. We summarize their properties as a proximity function for co-location mining as well.…”