MGsP: Extending the GsP to Support Semantic Interoperability of Geospatial Datacubes

摘要

Data warehouses are being considered as substantial elements for decision support systems. They are usually structured according to the multidimensional paradigm, i.e. datacubes. Geospatial datacubes contain geospatial components that allow geospatial visualization and aggregation. However, the simultaneous use of multiple geospatial datacubes, which may be heterogeneous in design or content, drives to consider interoperability between them. Overcoming the heterogeneity problems has been the principal aim of several research works for the last fifteen years. Among these works, the geosemantic proximity notion (GsP) represents a qualitative approach to measure the semantic similarity between geospatial concepts. The GsP, which has been defined in the transactional context, and can be used to a certain extent in the multidimensional paradigm, needs to be revisited to be more suitable for this paradigm. This paper proposes an extension to the GSP notion in order to support the semantic interoperability between multidimensional geospatial datacubes. The extension, called MGsP, aims to give the possibility to dig into and resolve semantic heterogeneity related to key notions of the multidimensional paradigm.