Towards Specialized Integrity Constraints for Spatial Datacubes

Spatial datacubes (also called "spatial multidimensional databases") are the cornerstone of the emerging Spatial On-Line Analytical Processing (SOLAP) technology. They are aimed at supporting Geographic Knowledge Discovery (GKD) as well as certain types of spatial decision-making. Although these technologies seem promising at first glance, they may provide unreliable results if one does not consider the quality of spatio-temporal data stored in them. In traditional spatial databases, spatial integrity constraints have been employed to improve internal quality of spatial data. However, for defining integrity constraints of spatial datacubes, these traditional spatial integrity constraints should be revisited. To this end, this paper presents the characteristics of spatial datacubes that differentiate them from transactional spatial databases from an integrity constraint perspective. These characteristics concern the datacubes model structure, the presence of thematic, spatial, and temporal data, and the various levels of data reliability for different decisions. Based on these characteristics, we propose spatial multidimensional integrity constraints and identify their fundamental characteristics. These characteristics include (1) considering the building elements of multidimensional data structures, (2) restricting thematic, spatial, and temporal data cross-tabulation, and (3) including a range of tolerance within the definition of integrity constraints. The analysis of today's solutions shows that existing spatial integrity constraint specification languages cannot express efficiently spatial multidimensional integrity constraints. Finally, future research directions for a formal model of spatial multidimensional integrity constraints is discussed as well as integrity constraints specification languages. * Corresponding author.