Understanding human dynamics is essential in predicting events that are potentially destructive to society. Analysis of publicly-available geospatial data reveals patterns of movement and trends in community activities that indicate baseline normal activity and provides potential opportunities to detect anomalies.
The recent growth of such publicly-available geospatial data sources now opens the possibility to document and understand the complicated dynamics of communities and identify patterns that may be indicative of threats. The ability to analyze such data in near real-time offers the potential to identify threat indicators quickly enough to prevent violent acts or limit their impact.
The Division of Mathematical Sciences (DMS) at the National Science Foundation (NSF) has a long history of supporting fundamental mathematical and statistical research relevant to the national interest. DMS has formed a partnership with joint proposal review and proposal management with the National Geospatial-Intelligence Agency (NGA) to develop the next generation of mathematical and statistical algorithms for the inference of information from large geospatial datasets.
The NSF’s Algorithms for Threat Detection (ATD) program supports research on new ways to use spatiotemporal datasets to develop quantitative models of human dynamics. The objectives include improved representation of complicated group dynamics and the development of algorithms that can process data in near real-time to accurately identify unusual events and forecast future threats indicated by those events.
The ATD program will support research projects that aim to develop novel mathematical and statistical algorithms for analysis of large geospatial datasets. Means to quantify confidence levels are desired, as are insights into new spatiotemporal datasets and valuable means of assembling them.
Models may range from those that address activities of individuals to those applicable to small groups or entire nations. These models may leverage mathematical research areas including, but not limited to, point processes, time series, dynamical systems, partial differential equations, and optimal control. Models that depend almost entirely on the spatial and temporal aspects of the data are of greatest interest. General applications of interest include threat detection, predictive analytics, human mobility, and human geography.
This program seeks ambitious and creative research proposals from individual investigators and collaborative groups in the mathematical sciences community. Research collaborations among mathematical scientists and social, behavioral, and economic scientists are encouraged.
The anticipated funding amount for this program is $3,000,000 annually, subject to the availability of funds. Further details are available by via NSF Program Solicitation: NSF 17-510. The next proposal deadline is February 21, 2017 and the third Tuesday in February, annually thereafter.