Big Data Analytics & GIS
Recent developments in sensor technology (e.g. earth observation, mobile sensors) have led to spatial data being collected in increasingly large amounts. At the same time, open access policies as in the European Copernicus programme further increase the volume of openly available data. To leverage the expanding availability of geospatial datasets, Geographic Information Systems (GIS) are being challenged to interface cloud-based processing and distributed storage, resulting in complex analytics infrastructures. Modern technologies such as Hadoop (see, e.g. esri.github.io/gis-tools-for-hadoop), Spark (e.g. GeoSpark, GeoTrellis), or array databases (e.g., rasdaman.com, paradigm4.com) have been successfully used for geoprocessing and demonstrate how these datasets can be handled in practice. For many users however, these tools are still difficult to use, and ideas how to facilitate the work with large datasets in GIS applications are still missing. A key challenge here is to provide highly scalable processing capabilities in combination with an easy to use GIS interface in order to make them available to a larger user group.
This workshop aims at bringing together developers and users from science and industry working with large volumes of geospatial data in Geographic Information Systems. Among others, the workshop will discuss:
- existing technical solutions to support scalable GIS analytics,
- specific requirements for processing various types of GIS data (e.g. satellite imagery, VGI, feature collections),
- typical applications of big GIS analytics in different domains,
- whether and where formal standards can be useful,
- how technology can be made accessible to GIS users with and without a strong technical background, and
- GIS-based visualisation of large datasets.