Efficient Discovery of Spatial Co-Location Patterns Using the iCPI-tree

Abstract

With the rapid growth and extensive applications of the spatial dataset, it's getting more important to solve how to find spatial knowledge automatically from spatial datasets. Spatial co-location patterns represent the subsets of features whose instances are frequently located together in geographic space. It's difficult to discovery co-location patterns because of the huge amount of data brought by the instances of spatial features. A large fraction of the computation time is devoted to identifying the table instances of co-location patterns. The essence of co-location patterns discovery and four co-location patterns mining algorithms proposed in recent years are analyzed, and a new join-less approach for co-location patterns mining, which based on a data structure-iCPI-tree (Improved Co-location Pattern Instance Tree), is proposed. The iCPI-tree is an improved version of the CPI-tree which materializes spatial neighbor relationships in order to accelerate the process of identifying co-location instances. This paper proves the correctness and completeness of the new approach. Finally, an experimental evaluations using synthetic and real world datasets show that the algorithm is computationally more efficient.

Partitioning Approach to Collocation Pattern Mining in Limited Memory Environment Using Materialized iCPI-Trees

“…However, this algorithm gives up the Apriori-like [1] strategy for generating new candidates, which in many cases can significantly reduce the number of candidates and therefore increase the overall performance. This problem has been addressed in [7]. The authors introduced a method based on a new structure called iCPI-tree.…”

“…It is represented as a branch of the iCPI-tree under the node with feature A. Sub-branches B and C determine neighbors with the appropriate spatial feature (B6 and C8,C11 respectively). See [7] for the iCPI-tree construction details.…”

“…In comparison with the join-based strategy, the number of potential collocation instances is dramatically reduced, however, some of them can form a star pattern instead of a clique, and therefore should be removed from the result set. In [7], the authors proposed a method that stores star neighborhoods in a tree structure. Modified procedure of generating candidates results in construction of proper collocation instances only.…”

“…Due to the complexity of the spatial data, internal data structures used by collocation discovery algorithms [4,6,7,9,11,12] can necessitate large amounts of the system memory. Nanopoulos et al consider in their work [5] real world database systems, where OLTP and data mining queries for association rules discovery are executed simultaneously.…”

Delaunay triangulation‐based spatial colocation pattern mining without distance thresholds

“…Thus, the performance of these algorithms degrades quickly when data sets are big or dense. Some algorithms that avoid this disadvantage as the Co‐location Pattern Instance‐tree [12] and Improved Co‐location Pattern Instance‐tree [13] algorithms were proposed. These algorithms directly generate table instances of patterns without the join operation by constructing a colocation pattern instance prefix tree.…”