A Data Cube Metamodel for Geographic Analysis Involving Heterogeneous Dimensions

Due to their multiple sources and structures, big spatial data require adapted tools be efficiently collected, summarized analyzed. For this purpose, are archived in warehouses explored by online analytical processing (SOLAP) through dynamic maps, charts tables. Data thus converted cubes characterized a multidimensional structure on which exploration is based. However, often lead several defined heterogeneous dimensions. In particular, dimensions definition can change depending analyzed scale, territory time. order consider these three issues specific geographic analysis, research proposes an original cube metamodel unified modeling language (UML). Based concepts like common dimension levels metadimensions, the instantiate constellations of allowing SOLAP perform multiscale, multi-territory time analysis. Afterwards, implemented relational warehouse validated operational tool designed for social economy case study. This tool, called “Racines”, gathers compares about business Belgium France interactive cross-border reports. Thanks metamodel, users remain independent from IT specialists regarding integration.