This thrust focuses on developing new methods for the analysis of publicly available data in order to anticipate and/or predict significant societal events, such as political instability, humanitarian crises, disease outbreaks, economic instability, and devastating effects of natural disasters. We aim to develop data assimilation algorithms, forecasting algorithms, and data collection platforms for studies on human behavior, with deep exploration into foundational issues of measurement, construct validity and reliability, and dependencies within data.
Quantifying the risk of local Zika virus transmission in the contiguous US during the 2015–2016 ZIKV epidemic
Microblog Conversation Recommendation via Joint Modeling of Topics and Discourse
Featured news coverage
CNN Tech, May 2017
The Data Forecasting Project focuses on developing new methodologies for forecasting by curating massive data sets from social media and mobility patterns. Epidemiological models will be built from tracking Infuenza-Like Illness (ILI) to detect early cases of ILI in small geographic regions; and from voter registration data collected from 1.7M Twitter handles from 86 countries of more than 500 elections. We use these data to develop a forecasting approach that combines digital indicators and mechanistic models. General formalizations of these forecasting models are applied to a wide range of behaviors including social movements, media consumption, and epidemiological prediction.