Modeling Spatial Dependencies for Mining Geospatial Data: A Statistical Approach

Abstract

Geo-spatial data mining is a process to discover interesting and potentially useful spatial patterns embedded in spatial databases. Efficient tools for extracting information from geo-spatial data sets can be of importance to organizations which own, generate and manage large geo-spatial data sets. The current approach towards solving spatial data mining problems is to use classical data mining tools after "materializing" spatial relationships and assuming independence between different data points. However, classical data mining methods often perform poorly on spatial data sets which have high spatial auto-correlation. This approach often leads to poor results because it does not take into account the fundamental notion of spatial autocorrelation. In this paper we will overview statistical techniques which can effectively model the notion of spatial-autocorrelation. We will also present a "roadmap" for extending classical data mining techniques to manage geo-spatial data which will the serve as basis for future research.