The Intelligence Advanced Research Projects Activity (IARPA) will host a Proposers’ Day Conference for the Mercury Program on March 15, 2015.
The Mercury Program seeks to develop methods for continuous, automated analysis of Signals Intelligence (SIGINT) in order to anticipate and detect political crises, disease outbreaks, terrorist activity, and military actions.
Past research has found that publicly available data can be used to accurately forecast events such as political crises and disease outbreaks. However, in many cases, relevant data are not available, have significant lag times, or lack accuracy. Little research has examined whether data from foreign SIGINT can be used to improve forecasting accuracy in these cases.
The Mercury Program will consist of both unclassified and classified research activities and expects to draw upon the strengths of academia and industry through collaborative teaming. It is anticipated that teams will be multidisciplinary, and might include social scientists, mathematicians, statisticians, computer scientists, content extraction experts, information theorists, and SIGINT subject matter experts with applied experience in the U.S. SIGINT System.
Anticipated innovation areas include:
- Processing and analysis of streaming data that represent significant population behavior changes.
- Development of data extraction techniques that focus on volume, rather than depth, by identifying shallow features of streaming SIGINT data that correlate with events.
- Development of models to generate probabilistic forecasts of future events.
Mercury will not fund research on U.S. events, or on the identification or movement of specific individuals, and will only leverage existing foreign SIGINT data for research purposes.
The conference will be held in the Washington, DC metropolitan area and provide introductory information on Mercury and the research problems that the program aims to address. The event will also provide a forum for potential proposers to present their capabilities and identify potential team partners.