WASHINGTON, Nov. 12, 2013 -- /PRNewswire/ -- The United States Agency for International Development (USAID) and Humanity United today announced the winners of the Model Challenge, the final round of the Tech Challenge for Atrocity Prevention. Run in collaboration with the Harvard-NASA Tournament Lab, the NASA Center of Excellence for Collaborative Innovation and TopCoder, the Model Challenge asked innovative problem solvers to create algorithmic models that can help forecast when and where mass atrocities are likely to occur. USAID and Humanity United hope that human rights organizations and governments can use the winning algorithms to create tools that make it easier to identify at-risk regions, leading to more effective atrocity prevention efforts.
Nearly 100 contributors submitted 618 algorithms for the challenge through TopCoder's online competition platform. The five most promising entries, all of which used historical events and sociopolitical data to accurately model the risk of violence in the near-term future, were awarded prizes ranging from $1,000 to $12,000.
Experts were particularly impressed with first-place winner Xiaoshi Lu of Beijing, China. His algorithm was especially successful at predicting atrocities in regions with limited or no past history of mass violence, taking into account 23 diverse geographic, socio-political and historical violence factors within each region. Lu was awarded $12,000 for his algorithm. The other winners were:
- Second Place ($6,000): Julian Labeit of Karlsruhe, Germany
- Third Place ($3,000): Zhanyi of Singapore
- Fourth Place ($2,000): Balázs Gődény of Veresegyhaz, Hungary
- Fifth Place ($1,000): Cyril Vincler of Reims, France
The judges also named an Ideation Winner, David Mace of the California Institute of Technology, who won $1,000 for his suggestion to evaluate the tensions between regions and the capacity for stress within a network as part of a model.
"In this final challenge, we sought innovative applicants that could access and apply newly available data to the prevention of mass atrocities," said USAID Deputy Assistant Administrator Sarah Mendelson. "We were pleased with the entries we received, and we are optimistic that the winning solutions will contribute to the development of new tools to support decision makers in the future."
Since its launch in October 2012, the Tech Challenge has issued five sub-challenges related to atrocity prevention. With the final sub-challenge now complete, Humanity United and USAID will explore ways to pilot and scale the most promising innovations.
"Complex challenges such as preventing atrocities require innovative solutions," said Randy Newcomb, president and CEO of Humanity United. "What's exciting about this challenge is the application of advanced analytics to a diverse set of data, to help us better understand where atrocities might occur next. We look forward to working with the winners to refine and expand these ideas."
The Model Challenge drew from two public datasets: the Global Database of Events, Language, and Tone (GDELT) and the Political Instability Task Force (PITF) data on worldwide atrocities.
More detail on each of the challenges and winners is available at http://thetechchallenge.org/winners.html.