Lockheed Martin’s Integrated Crisis Early Warning System (ICEWS) Program Records 10,000 Dataset Download

Lockheed Martin Advanced Technology Laboratories’ (ATL) Integrated Crisis Early Warning System (ICEWS) team recently achieved a major program milestone – the program recorded 10,000 downloads of their dataset from the Harvard Dataverse.

What's a Dataverse?

The Harvard Dataverse Network is an open-source application for sharing, citing, analyzing and preserving research data. It includes the world's largest collection of social science research data and is maintained by Harvard University’s Institute for Quantitative Social Sciences.

10,000...That's a Lot of Downloads

The ICEWS dataset has been available on the Harvard Dataverse since June 2015, allowing researchers to draw correlations among past events to help forecast future events. Knowing it’s been downloaded 10,000 times means it’s making a significant impact on the academic community, giving University researchers powerful data analytics to help forecast future unrest and anticipate the next uprising or protest.

What is ICEWS?

ICEWS is a sophisticated database system that captures and processes millions of pieces of data from digitized news media, social media and other sources to predict, track and respond to events around the world. It combines modeling with situational understanding to provide U.S. government agencies with the ability to monitor, assess and forecast crises around the world. This system can help organizations understand the past and present as well as anticipate future challenges and help mitigate those associated risks.

Helping the Academic Community

ICEWS was also one of four political event datasets required and recommended for use by participants in the 2016 Grand Data Challenge at the International Conference on Social Computing, Behavioral-Cultural Modeling & Prediction and Behavior Representation in Modeling and Simulation (SBP-BRiMS2016). This year’s challenge problem asks participants to address any issue of interest that involves events and their distribution over time or space.

The ICEWS team uploaded more than 15 million events extracted from news stories onto the Harvard Dataverse to share with the research community. Approximately 50,000 new events are uploaded monthly, having been sanctioned by the U.S. Government for public release.

For more information, contact icews@atl.lmco.com

This work was funded, in part, by the Office of Naval Research, contract #N00014-14-C-0137.