Article Preview
TopSpatial Data Mining
In recent years geographic data collection devices linked to location-aware technologies such as the global positioning system allow researchers to collect huge amounts of data. Other devices such as cell phones, in-vehicle navigation systems and wireless Internet clients can capture data on individual movement patterns. This explosive growth of spatial data and widespread use of spatial databases emphasize the need for the automated discovery of spatial knowledge.
The process of extracting information and knowledge from these massive geo-referenced databases is known as Geographic Knowledge Discovery (GKD) or Spatial Data Mining. It may be useful to understand spatial data, to discover relationships between spatial and non spatial data, to build knowledge-bases. This has a wide application in Geographic Information Systems (GIS), image analysis and other different areas where spatial data are used.
The nature of geographic entities, their complexity, relationships, and data means that standard Knowledge Discovery in Databases (KDD) or Data Mining techniques are not sufficient (Koperski, 1998) or at least their usefulness is limited. In fact the data inputs of Spatial Data Mining include extended objects such as points, lines, and polygons.
Specific reasons are the nature of geographic space, the complexity of spatial objects and relationships as well as their transformations over time, the heterogeneous and sometimes ill-structured nature of geo-referenced data, and the nature of geographic knowledge.