| Spatial analysis is the quantitative study of phenomena that are located in space. Spatial analysis encompasses a wide range of techniques for computing and analyzing geographic data. This thesis presents an examination of efficient processing techniques for spatial analysis. We address spatial processing techniques in three particular contexts, namely, spatial query processing, spatial data warehouse, and spatial data mining.; In spatial query processing, we focus on the problem of efficient processing of spatial join. A join-index is a data structure used for processing join queries. Join-indices use pre-computation techniques to speed up online query processing. Given a join-index, we introduce a suite of methods based on clustering to compute the joins. We derive upper-bounds on the lengths of the page-access sequences. Experimental results with Sequoia 2000 data sets show that the clustering method outperforms existing methods based on sorting and online-clustering heuristics.; A data warehouse is a repository of subject-oriented, integrated, and non-volatile information. A data cube operator is used to generate the union of a set of alpha-numeric summary tables corresponding to a given aggregation hierarchy. We extend the concept of a data cube to the spatial domain by proposing “map cube,” an operator which takes a base map, associated data tables, cartographic preferences, and generates an album of maps. A map cube organizes the album of generated maps using the given aggregation hierarchy to support browsing via roll-up, drill-down, and other operators on aggregation hierarchy. We use 1990 census data to illustrate the notion of map cube, and discuss research issues raised by the map cube operator.; In spatial data mining, we focus on spatial outlier detection, which can lead to the discovery of unexpected and important knowledge. We provide a general definition of S-outlier which subsumes the traditional definitions of spatial outliers. The structure for S-outlier detection computation is characterized. We propose a scalable algorithm to detect S-outliers, and analyze the algebraic cost model for the proposed algorithm. In addition, we provide experimental evaluation of our algorithm using a real-world data set describing traffic on Minneapolis-St. Paul highways. |
