A New Architecture for Geographical Knowledge Discovery Based on Topological Dimensions and Measurements

Data Warehouse (DW) and Online Analytical Processing (OLAP) systems based on multi-dimensional data models include dimensions and measurements. These systems are used for data mining and knowledge discovery processes. Geographical Data Warehouse (GDW) systems are able to store and represent the geographic data cartographically but they can not do temporal and topological analysis, which are very important for geographical knowledge discovery. In this paper, we define and implement a new set of dimensions and spatial measurements based on 4-intersection relationship model between geographical objects, called topological dimensions and measurements. This proposed multi-dimensional model is implemented using SpatialOracle9i software package. We have evaluated its ability to respond to analytical queries of users. Our experiments show that the designed GDW can answer all kind of analytical queries and express topological relationships between two geographical objects.