Data Science for Social Good projects begin

By Emily Keller-O’Donnell

July 9, 2019

The 2019 Data Science fellows pose for a group shot. Photo, Robin Brooks, eScience Insitute
The 2019 Data Science fellows pose for a group shot. Photo, Robin Brooks, eScience Institute

On June 17th, the Data Science for Social Good (DSSG) summer program began with 15 fellows working on four project teams in collaboration with data science leads from the eScience Institute and project leads from academia and government. Student fellows, including five international students, joined the program from colleges and universities around the country. Their areas of study range from sociology, public policy and economics to oceanography, statistics and computer science.

This marks the fifth year of the program that matches data scientists with a diverse range of students from around the world to tackle real-world social issues through multi-disciplinary, collaborative work. Project leads from academia, public agencies, non-profits and industry guide teams through the ten-week program, which takes place at eScience Institute’s Data Science Studio. Each team also participates in systematic and sustained engagement with an array of stakeholders related to their projects.

The 2019 projects are described below:

“ADUniverse: Evaluating the Feasibility of (Affordable) Accessory Dwelling Units in Seattle” examines regulatory, financial, design, and permitting issues around building homeowner-developed accessory dwelling units (ADUs) to increase affordable housing in Seattle. The project is led by Rick Mohler, a University of Washington (UW) Associate Professor and member of the Seattle Planning Commission, and Nick Welch, a Senior Planner at the Seattle Office of Planning and Community Development. Deliverables will include a citywide feasibility analysis and a prototype web-based interactive tool that estimates suitability for an ADU on individual properties.

“Developing an Algorithmic Equity Toolkit with Government, Advocates, and Community Partners” is led by Mike Katell, a PhD candidate at the UW Information School. In partnership with the American Civil Liberties Union (ACLU) of Washington, the project will develop a set of tools for identifying and auditing public sector algorithmic systems, with a focus on automated decision-making and predictive technologies. The tools are intended to serve state and local government employees, and advocacy and grassroots organizations.

“Understanding Congestion Pricing, Travel Behavior, and Price Sensitivity,” led by Mark Hallenbeck, Director of the Washington State Transportation Center at UW, utilizes data from I-405’s congestion priced Express Lanes. In partnership with the Washington State Department of Transportation, this project will determine the time savings achieved by Express Lanes users; variations in time savings based on facility price; how socio-demographic, geographic, and mode choice factors affect the distribution of costs and benefits of the Express Lanes; and the impact of user behavior on roadway performance.

“Natural Language Processing for Peer Support in Online Mental Health Communities” is led by UW Assistant Professor Tim Althoff and UW Research Professor Dave Atkins. The project utilizes a data set of 100 million posts and interactions through an online peer support platform, to identify the types of communications that are most helpful to young adults who seek support online.  The results will be used to develop tools and trainings for peers who want to support others in need.

View detailed project descriptions here. The program will conclude with project presentations to a public audience on Wednesday, Aug. 21 from 3 to 5 p.m. Additional information will be available later this summer.