Skip to content

Forecasting International Events

Photo of
Hosted By
Harlan H. and 3 others


For (late) April, Data Science DC is very pleased to welcome two speakers bringing their perspectives and technical approaches to forecasting events on the international stage. Prof. Naren Ramakrishnan and Dr. Jay Ulfelder will talk about how they've tackled the problem of integrating diverse, noisy data sets and used statistical and machine learning tools to forecast events such as disease outbreaks and political protests. If you're looking to learn about real-world data science, it doesn't get any more real-world than this.


• 6:30pm -- Networking, Empanadas, and Refreshments

• 7:00pm -- Introduction, Announcements, Give-aways

• 7:15pm -- Presentations and Discussion

• 8:30pm -- Data Drinks (Tonic, 2201 G St.)


Forecasting Atrocities

The Early Warning Project ( aims to help prevent mass atrocities around the world by producing and publicly disseminating earlier and more accurate forecasts of them. At present, the project uses two approaches: statistical risk assessments and aggregated expert judgment. This talk will focus on the statistical component, which uses publicly available country-year data to produce annual assessments of the risk of new episodes of state-led mass killing -- a very rare event -- in countries worldwide. Those assessments are generated with a multi-model ensemble that combines parametric statistical models with Random Forests.

Jay Ulfelder is a political scientist whose research interests include democratization, political violence, social unrest, state collapse, and forecasting methods. He received his Ph.D. in from Stanford University in 1997 and his B.A. in Comparative Area Studies (USSR and Eastern Europe) from Duke University in 1991. From 2001 to 2011, Ulfelder served as research director for the Political Instability Task Force, a U.S. government-funded research program that develops statistical models to forecast various political events around the world. He now makes a living as a consulting researcher and has spent most of the past three years working with the U.S. Holocaust Memorial Museum's Center for the Prevention of Genocide to develop the Early Warning Project (, a public early-warning system for mass atrocities around the world. His blog, Dart-Throwing Chimp (, won the International Studies Association's Best Blog (Individual) Award in 2014 and 2015. Follow Jay on Twitter @dtchimp (

Beating the news with EMBERS: Forecasting Significant Societal Events using Open Source Indicators

We describe the design, implementation, and evaluation of EMBERS, an automated, 24x7 system for forecasting significant societal events such as disease outbreaks, civil unrest, and elections in countries of Latin America. EMBERS ingests a broad variety of open source datasets such as tweets, news, blogs, economic indicators, and distills them into forecasts. Unlike retrospective studies, EMBERS has been making forecasts into the future since Nov 2012 which have been evaluated by an independent team. Of note, EMBERS has successfully forecast the June 2013 protests in Brazil, the Feb 2014 violent protests in Venezuela, and Hantavirus outbreaks in Chile and Argentina. We outline the system architecture of EMBERS, individual models that leverage specific data sources, and a fusion and suppression engine that supports trading off specific evaluation criteria.

Naren Ramakrishnan ( is the Thomas L. Phillips Professor of Engineering at Virginia Tech in Arlington, VA, and director of the university's Discovery Analytics Center. His work has been featured in many venues including the Wall Street Journal, Popular Science, Newsweek, The Chronicle of Higher Education, and Smithsonian Magazine. He received his PhD in Computer Sciences from Purdue University.


This event is sponsored by the GWU Department of Decision Sciences (, (, Elder Research (, Novetta Solutions (, Booz Allen Hamilton (, and Pearson/InformIT ( (Would your organization like to sponsor too? Please get in touch!)

1957 E Street · Washington, DC
83 spots left