GeoTimeLine - A system for multi-hazard event reconstruction in Space and Time
Description
Information adequately documented is crucial in supporting decision-making. The expansion of Web 2.0 made available loads of information, with a great potential in a wide range of domains, and opened up to authorities a series of opportunities towards better decisions. This fact raises a few issues though related to vast amounts of unclassified data, various sources and formats, and their dispersion across the Web. Thus, one of the main challenges refers to the ability of effectively mining, validating, annotating, and integrating key data to extract relevant information about a specific context of interest, especially within very short time windows. As far as our own research interests are concerned, we are particularly interested in enriching data related to a specific event by inferring additional details on both their geo-location (where) and time-location (when). In the light of the above, the main aim of this project is to establish documented databases provided with a “geotimeline” creation mechanism. Types of data to search and to be downloaded onto the database include: 1) Crowdsourced data gathered by citizens for various purposes (e.g. social networks); 2) Environmental data from in-situ physical sensors (e.g. water gauges); 3) Online newspaper archives. The integration of diverse sources of such data within a single enriched database provided with features above will enable to: 1) Produce geotimelines of historical key events that took place and show both the geospatial and temporal organisational evolution of those events; 2) Set up a stand-alone system whose data validation & annotation tools, and geotimeline creation feature can be integrated into other decision support platforms (e.g. for risk/emergency mitigation, our main immediate goal). The development of such system will improve the support of decision making processes may assist by providing authorities with valuable information, both in quantity and quality, especially in real-time situations where decisions directly affecting citizens need to be taken rapidly.Researchers
José Paulo Duarte de Almeida (coordinator)
Alberto Cardoso
Cidália Fonte
Jacinto Estima
Joaquim Patriarca
Alberto Cardoso
Cidália Fonte
Jacinto Estima
Joaquim Patriarca