Virginia Tech is addressing the growing need for data scientists to help answer complex questions of urban populations by offering a 12-credit graduate certificate in urban computing.

The program, funded with a grant of nearly $3 million over a five-year period from the National Science Foundation Research Traineeship Program, is available to master's and Ph.D. students in Blacksburg and the National Capital Region in fall 2017.

“As cities become more wired and networked, the need for scientists who can analyze large-scale operational and behavioral datasets has grown,” said Naren Ramakrishnan, the Thomas L. Phillips Professor of Engineering in the Department of Computer Science and director of the Discovery Analytics Center. “Projections show that by 2030, six out of every 10 people will live in a city, and by 2050, seven out of 10.  Key issues, such as public health, sustainable use of limited energy resources, emergency preparedness, and societal stability, are rising to the forefront.”

Administered through the Discovery Analytics Center, the program weaves interdisciplinary applications through new courses and a novel “tapestry” curriculum.

These courses are designed to train students to become competent problem solvers by developing computational models of urban populations from disparate data sources and posing and answering what-if questions via machine learning and visualization methodologies. Students are also trained in the ethical and professional implications of working with massive datasets. 

In the tapestry curriculum, students choose courses from a set of "horizontals" (covering foundations of data science) and a set of "verticals" (covering applications to urban populations) to tailor their graduate plan of study. In addition, faculty representing at least two different departments co-advise every urban computing thesis/dissertation.

While the certificate program is formally launching in the fall, a handful of students already are pursuing the certificate, working on urban computing projects supported by the National Science Foundation grant. Following are a few examples of the kinds of research opportunities the program offers:

  • Location, neighboring property values, hedonic measures of a specific home, and changes in the overall housing market can all affect housing prices. Matt Slifko, of Johnstown, Pennsylvania, a Ph.D student in statistics, is building a spatiotemporal model for housing prices in the United States. Ultimately, he will use the model to determine how knowledge about the housing market in one area translates into knowledge regarding other areas. Slifko’s co-advisors are Scotland Leman, associate professor of statistics, and David Bieri, associate professor of urban affairs and planning.
  • Capacity drops that produce additional traffic congestion are a critical problem in urban transportation networks. Huthaifa Ashqar, of Athens, Ohio, a Ph.D. student in civil engineering, is developing a speed harmonization (SH) algorithm based on a vehicle-to-infrastructure (V2I) communication system to test as a means of regulating the flow of approaching traffic. His co-advisors are Hesham Rakha, the Samuel Reynolds Pritchard Professor of Engineering, and Leanna House, associate professor of statistics.
  • Electricity infrastructure is becoming “smarter” in today’s urban environments with the deployment of micro grids that use renewable energy sources and distributed generation. To quickly identify anomalies and prevent catastrophic failure of the power system, critical infrastructure needs constant monitoring. Nikhil Muralidhar, of Fairfax, Virginia, a Ph.D. student in computer science, is working on a real-time anomaly detection and state estimation system for power systems with a focus on micro grids. His co-advisors are Ramakrishnan and Saifur Rahman, the Joseph R. Loring Professor of Electrical and Computer Engineering and director of the Advanced Research InstituteNational Capital Region.
  • Some modern buildings are equipped with accelerometers that can "hear" vibrations in the floor. Most vibrations are caused by ordinary building activity like air conditioning and people walking, but vibrations are also caused by problems like footsteps in restricted areas or someone tripping and falling. Jonathan Baker, of Huntsville, Utah, a Ph.D. student in math, is developing techniques to recognize automatically the vibration of "sound" in an emergency. His co-advisors are Mark Embree, professor of mathematics, and Pablo Tarazaga, assistant professor of mechanical engineering. 

Ramakrishnan said that the urban computing certificate was designed with input from faculty in Virginia Tech’s computer science, statistics, mathematics, electrical and computer engineering, civil and environmental engineering, population health sciences, sociology, and urban affairs and planning graduate programs. However, the certificate is not limited to students in these programs. Students in other graduate programs at Virginia Tech are not precluded from pursuing the certificate.

For more information, contact Wanawsha Hawrami at the Discovery Analytics Center.

Share this story