Newswise — Sniffing out corruption in World Bank contracts. Predicting where Chicago children are most at risk for lead poisoning. Identifying struggling high school students who need additional guidance. Reducing maternal mortality in Mexico.
This summer, 48 fellows from around the world have come to Chicago to tackle these and other missions as part of the 2014 Data Science for Social Good Summer Fellowship, a University of Chicago program dedicated to making a better world through innovative data science projects. For 12 weeks, fellows work with non-profit and government organizations to make the most of their data and create positive impact in education, healthcare, energy conservation, economic development and more.
“The primary goal for this fellowship is to get fellows trained in both solving real problems and solving problems with real impact,” said Rayid Ghani, director of the Data Science for Social Good Summer Fellowship. “For this year, we’ve been able to build better partnerships and build projects that go deeper and help us and our partners solve their problems in a better way.”
This year’s fellows were chosen from over 300 applications received from two dozen countries and over 100 universities. Undergraduate and graduate students specializing in computer science, statistics, public policy, economics, and other areas are organized into teams with experienced mentors from academia and industry, working with partner organizations from non-profit and government sectors.
Partners for the 2014 program include The World Bank Group, the Office of the President of Mexico, Chicago Public Schools, Sunlight Foundation, the Chicago Department of Public Health, Enroll America, Get Covered Illinois, the City of Memphis, Nurse-Family Partnership, Health Leads, the Chicago Alliance to End Homelessness, Montgomery County Public Schools, Pecan Street, Conservation International, and Skills for Chicagoland’s Future.
In the project with the Chicago Department of Public Health, DSSG fellows will use data on blood tests, property records, demographics, and other sources to predict buildings at risk for lead contamination and likely to house children or pregnant mothers. Their model will help CDPH more efficiently deploy inspections and lead mitigation efforts before harm is done to Chicago children.
“Engaging predictive analytics in public health is an innovative way to leverage data to generate valuable insights that are actionable to improving health,” said Dr. Bechara Choucair, Commissioner of the Chicago Department of Public Health. “With partners like Data Science for Social Good on the lead predictive pilot, public health is at the cutting edge of improving the lives of Chicagoans.”
Another project, with Montgomery County Public Schools in Maryland, builds upon work last year by DSSG fellows to predict and prevent student “under-matching” -- when a student enrolls in a college below their full potential. With student data from Montgomery County, fellows are building an “early warning” model to identify students who are not making sufficient academic progress, so that schools can intervene and put them back on track.
In addition to helping partners do more with their data and building powerful new open-source tools, the program provides fellows with valuable experience working with real-world organizations outside of the technology industry. The goal is not just to complete innovative and transformative projects this summer, Ghani said, but to inspire a new generation of aspiring data scientists to continue to work on more meaningful projects that benefit the world.
“It’s great to see a community building, and that’s what we really need in the long term,” Ghani said. “It’s not just the projects happening here, but also the fellowship seeding a larger community that wants to use data science skills to help society.”
Data Science for Social Good is a University of Chicago program co-administered by the Computation Institute and the Harris School of Public Policy. For more information, visit dssg.uchicago.edu