Home > Launching Data Science & Machine Learning Prize Competitions
Launching Data Science & Machine Learning Prize Competitions
CLIENT
National Geospatial Intelligence Agency (NGA)
INDUSTRY
- National Defense
01
CASE
Machine Learning, Artificial Intelligence (AI), and Data Science Prize Competitions for the Intelligence Community
02
Overview
The National Geospatial-Intelligence Agency (NGA) is the nation’s primary source of geospatial intelligence (GEOINT). NGA sought novel approaches to solving data science problems. Ensemble, in partnership with Topcoder and Capital Consulting Corporation, launched two specific data science prize competitions seeking unique algorithms to address two key machine learning capabilities in computer vision and acoustical analysis.
In particular, NGA tapped the Topcoder crowdsourcing community to develop data science solutions that 1) detect, delineate, and describe circular-shaped objects in satellite imagery and 2) geo-locating video and audio recordings based on non-speech ambient sound.
03
OUR
Approach
We helped create and execute two prize challenges for the National Geospatial-Intelligence Agency (NGA) for crowdsourcing unique data science solutions to modern-day geospatial analysis problems. We led the integration of the Topcoder platform where the challenges were hosted. Alongside Topcoder, we helped in defining challenge components within the challenge platform, such as wireframes and storyboards for front-end design, code and module assembly challenges for development, and “marathon” leaderboard driven challenges for data science. We also led all communications and outreach efforts, including social media, webinar hosting, and earned media coordination.
Problem statements were established with feedback from NGA and likely solvers. Additionally, we supported problem statement refinement throughout the sprint cycle with project stakeholders, resulting in salient and compelling problem statements that can be solved by the crowd.
We developed a comprehensive communication plan detailing the purpose and subject of the competition including a Social Media outreach strategy, webinar, and “influencer kits” for engaging data science thought leaders to promote the competitions on behalf of the NGA.
05
Results
Website had 761 monthly visitors 10/2020
Majority of traffic to site was a result of direct or social searches
147 active weekly users
Demographics: 18-34. (27.5% 18-24, 33.5% 25-34)