Um framework para a análise espacial de fontes de dados multidimensionais.

Abstract

Business Intelligence (BI) and Geographic Information Systems (GIS) technologies have been used by several organizations to extract information from multiple data sources in order to improve the process of decision making. Several solutions for the integration of these two technologies have been proposed in recent years. The term Spatial OLAP arose from these attempts of integration and is now used to classify the tools to perform analysis on spatial and conventional data organized according to the multidimensional approach. Nevertheless, no consensus has been reached regarding the best way to accomplish such integration, making it difficult to perform analysis with spatial cubes from various multidimensional data sources in the same environment. In this work, we realize a study on the main characteristics of the state of the art in SOLAP, and we propose an environment that enables the analysis of spatial cubes from different multidimensional data servers. The proposed framework also has a visual query language for the spatial analysis. To validate the proposed environment, a practical example is conducted and applied to accountability processes of the Court of Auditors of the Acre's State, in Brazil. To perform such case study, the framework was extended to access cubes from Microsoft SQL Server Analysis Services (SSAS) and GeoMondrian.