João Moura Pires (superv.), Universidade Nova de Lisboa, October 2015.
Keywords: Spatio-temporal Data, Information Integration, Multidimensional Modeling
Abstract: In the current era of information a lot of geo-referenced data is produced and made available by many suppliers, on a national and international level, by public or private institutions and organizations with our without objectives of profit. Some of them are the primary source of the data while others consist of aggregators of data from various sources. The geographic and temporal level of detail of the data ranges broadly, from information related to a single coordinate in space and an instant in time, to data associated with a bounded region and a large time period. The data catalogs supplied by each source are also very heterogeneous when considering their quality, format, description capabilities and functionality.
The ability to effectively combine geographical and temporal information in maps and other analytical visualization tools strongly depends on our knowledge of the pertinence, availability and spatio-temporal frame of the information. From the analytical geo-visualization standpoint this ability is key to provide the user with proper analysis capabilities.
This dissertation provides an information model that supports an integration and query platform for spatio-temporal measures of phenomena. This information model exposes the relations that exist among these geographical and temporal measures through their semantic meaning, geographical and temporal coverage, level of detail, provenance, units of measure and by the different combinations and aggregations that can possibility be applied to the data.
Dissertation and presentation will soon be available.
Demonstration video (in Portuguese)
XML Schema of the meta model
Instantiation examples of the meta model