Leveraging the power of spatial data mining to enhance the applicability of GIS technology

The strength of a geographic information system (GIS) is in providing a rich data infrastructure for combining disparate data in meaningful ways, by using a spatial arrangement (e.g., proximity). As a toolbox, a GIS allows planners to perform spatial analysis using geo-processing functions, such as map overlay, connectivity measurements, or thematic map coloring. Although this makes the geographic visualization of individual variables effective, complex multi-variate dependencies are easily overlooked. The required step to take GIS beyond a tool for automating cartography is to incorporate the ability of analyzing and condensing a large number of geo-referenced variables into a single forecast or score. This is where spatial data mining promises great potential benefits and the reason why there is such a hand-in-glove fit between GIS and data mining facilities. INGENS 2.0 is a prototype GIS which resorts to emerging spatial data mining technology to support geographers, geologists, and town planners in discovering (descriptive and predictive) patterns, which are never explicitly represented in topographic maps or in a GIS-model and are useful in the task of topographic map interpretation. In spatial data mining, spatial dimension adds a substantial complexity to the data mining task. First, spatial objects are characterized by a geometrical representation and relative positioning with respect to a reference system, which implicitly define spatial properties. Modeling these implicit spatial properties (attributes and relations) in order to associate them with clear semantics and a set of eficient procedures for their computation is the first challenge to be met when facing a spatial data mining problem. Second, spatial phenomena are characterized by autocorrelation, i.e., observations of spatially distributed random variables are not location-independent. Third, spatial objects can be considered at different levels of abstraction (or granularity). Spatial data mining facilities in INGENS deal with these challenges in both inducing classification rules and discovering association rules from spatial data. The spatial data mining process is aimed at a user who controls the parameters of the process by means of a query written in SDMOQL, a spatial data mining query language that permits the specification of the task-relevant data, the kind of knowledge to be mined, the background knowledge and the hierarchies and the interestingness measures. Some constraints on the query language are identified by the particular mining task. An application to a real repository of topographic maps is briefly illustrated