MIDAS collaborates with community partners and supports the application of data science into impactful products, services, and policies, towards a better data-enabled society. Such collaboration also ensures that data science research is inspired by real-world problems and rooted in real-world data. Our researchers and students help community partners define research questions, design data strategies, and carry out data analytics. Below is a sample of current projects.
Reducing Vehicle Emissions and Improving Air Quality
In collaboration with the US Environmental Protection Agency (EPA) National Vehicle and Fuel Emissions Laboratory (NVFEL), MIDAS offers project-based classes. MIDAS Senior Scientist Jonathan Gryak teaches the classes in which students apply data science techniques to datasets provided by the EPA, with the goal of reducing the environmental impact of personal and freight mobility systems, designing systems to reduce vehicle emissions using connected and autonomous vehicles, and analyzing the potential environmental benefits of shared and/or automated vehicles.
Improving Transportation Systems in Detroit
The Michigan-Data Informed Cities for Everyone (M-DICE) research team is working with the City of Detroit to provide a unified solution for managing and integrating a large number of transportation data sets, and to use data to inform budget planning, resource allocation, and policy evaluation. MIDAS affiliate faculty member and Assistant Professor of Computer Science and Engineering, Danai Koutra, and MIDAS data science fellow, Arya Farahi, lead a group of students for this project which was highlighted as one of the World Economic Forum’s key future directions.
“I’ve always been fascinated by the complexity of transportation systems and their effect on society, and through my training in physics I’ve cultivated a set of analytical tools that I felt could allow me to provide a unique perspective on how these systems are organized. The Detroit Transportation project was a great opportunity for me to engage with real world problems and policy makers to directly impact change with these methods.” — Alec Kirkley, student
Improving Databases of Scientific Literature
How many scientists do you know are “J. Chen”, or “H. Kim”? Do you know “Rachmaninoff” and “Rachmaninov” can refer to the same person? Does your algorithm place “Computer Science Department” and “Department of Computer Science” in one category or two? Disambiguating names is essential in developing an accurate database in many areas of research. MIDAS affiliate faculty member, Jinseok Kim, works with the American Mathematical Society to apply state-of-the-art data science techniques to automatically disambiguate author names of millions of publications in mathematical journals. This work will greatly improve the efficiency of literature search and allow us to understand the landscape of math research. This is one example of how MIDAS collaborates with non-profit organizations to transform traditional research with data science.
Evaluating the Impact of Youth Programs
The Detroit Police Athletic League (PAL) has been offering athletic programs since 1969, to Detroiters ages 4-19 from low-income, under-served families. MIDAS affiliated faculty and Research Associate Professor at the Survey Research Center, Brady West, leads the effort to analyze survey data to assess the impact of the PAL programs on participants and their families, and how these programs improve the relationship between the police force and the community. The project is funded through a generous donation from Mark and Eileen Petroff. “Being an employer located in Detroit, I am so excited to see the outcome of this work”, says Mark, a U-M alumnus and the President and CEO of OneMagnify.