We present a social media crisis mapping platform for natural disasters. We take locations from gazetteer, street map and volunteered geographic information (VGI) sources for areas at risk of disaster and match them to geo-parsed real-time tweet data streams. We use statistical analysis to generate real-time crisis maps. Geo-parsing results are benchmarked against existing published work and evaluated across multi-lingual datasets. We report two case studies comparing 5-day tweet crisis maps to official post-event impact assessment from the US National Geospatial Agency (NGA) compiled from verified satellite and aerial imagery sources.

The increasing success of wireless technologies is sustaining the diffusion of mobile information systems, but the youth of the underl3äng technology and its peculiar characteristics are impacting the development of such systems. For example, the execution of business processes in such a context must cope with the variable and fluctuating bandwidth available to the different devices. This leads the designer to stress the independence of each actor ~ by minimizing interactions and knowledge sharing-to increase the reliability of the whole system. To this end, the paper proposes a rigorous approach for partitioning the execution of BPEL workflows on sets of portable devices, that is, the infrastructure of mobile information systems. The approach abstracts BPEL processes into attributed graphs and uses a graph transformation system as rules to split single workflows into meaningful sets of related processes. The paper presents such rules and exemplifies them on a case study in the cultural heritage domain.

No abstract

Abstract. This paper motivates the enablement of the Future Internet to become a highly functional service platform supporting the design and the operation of software applications in the Environmental Information Space. It reports on the experience made by the European research project ENVIROFI as one of the usage area projects within the Future Internet Public-Private Partnership programme. It describes the software components (environmental and specific enablers) which are required to connect with the domain-independent capabilities (generic enablers) of the Future Internet core platform for geospatially and environmentally-driven applications.

No abstract