| Popis: |
Analysis of 9-1-1 call stream data will provide a better understanding of the spatiotemporal patterns of emergency calls, both State-wide and at the local level, and their correlation with external events. Predictive models built with this data can lead to real-time decision support and better overall planning to enable more efficient and effective response to emergencies. While 9-1-1 data is currently being collected across the nation, it is being used primarily for administrative purposes, and not for real-time assessment and prediction of emergency response situations. The objective of this research is to provide that linkage between the data and the spatiotemporal analysis techniques required to mine the data and develop predictive models. This analysis will enable us to detect regular trends versus "unusual" events in the call stream data and establish spatiotemporal patterns and trajectories. Our analysis will include correlation of this information with external event information (e.g. earthquakes and wildfires) to determine spatiotemporal "signatures" of such events. This information then will be used to establish alarm thresholds to provide advanced warning to PSAP's, first responders, and other emergency service personnel, concerning the spatial extent and temporal evolution of an emergency event. Our long-term goal is also to enable the use of this information for real-time detection of emergency events in order to provide rapid response at the local level and facilitate decision support for resource allocation and planning at the State level. |
