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 Signals Intelligence (SIGINT) can be used to improve forecasting accuracy: communications between people or other electronic signals not directly used in communication.
The IARPA Mercury Program seeks to develop methods for continuous, automated analysis of SIGINT in order to anticipate and/or detect political crises, disease outbreaks, terrorist activity, and military actions. Successful proposers will combine cutting-edge research with the ability to develop robust forecasting capabilities from SIGINT data.
Anticipated innovations include: development of empirically driven sociological models for population-level behavior change in anticipation of, and response to, these events; processing and analysis of streaming data that represent those population behavior changes.
Also development of data extraction techniques that focus on volume, rather than depth, by identifying shallow features of streaming SIGINT data that correlate with events; and development of models to generate probabilistic forecasts of future events.