Win tickets to see the smash hit musical Mamma Mia at the Roanoke Civic Center. Two winners will each receive four tickets!
The proposed institute would be able to analyze data streams collected by different researchers.
Saturday, August 24, 2013
The University of Virginia is kicking off a research initiative on data analysis and application that it hopes will be a model for future research institutes at the university.
The Big Data Institute is a collaboration between different departments at the university that looks at ways to use what Rick Horwitz, UVa associate vice president for biosciences, calls the “overwhelming” streams of data collected by the government, corporations and academic researchers.
The institute has a prominent role in UVa’s new strategic plan because, administrators say, it’s an example of a low-cost research initiative that’s timely and immediately practical.
The project also will encourage interdisciplinary work, something the university wants to do more of moving forward.
Tom Skalak, vice president for research at UVa, said the university is just the right size for this kind of collaboration.
“We feel UVa can do something universities that are much larger couldn’t do because their units are more separate,” Skalak said.
The institute will look at ways to organize and make sense of the massive streams of data coming in, Skalak said. For example, astronomers and chemists are using telescopes to look at the chemical signals of other planets, searching for the presence of chemicals that could nurture, or perhaps even indicate, life. To analyze the data coming in, he said, would take “terabytes a day or more.”
“It takes a collaboration of astronomers, chemists and computer scientists who can design systems to handle streams of data,” Skalak said.
He cited a couple of other uses for data research: Data scientists could work with doctors to calculate patients’ risks of disease or adverse reactions to certain treatments. And they could work with businesses to find more effective ways to advertise.
Skalak said departments around the university have been talking about the need for more data analysis for the past couple of years.
“We really see it as a bubbling up from the grassroots across all schools,” he said. “Many people in these many disciplines felt like there was an onslaught of data in every field.”
Data research institutes are not new. The University of North Carolina at Chapel Hill, Columbia University and the Massachusetts Institute of Technology are just a few schools with similar initiatives. But UVa wants to leverage its traditional strengths in social sciences and the humanities into certain niches in the field.
For example, Horwitz said, the university could look into the legal and ethical implications of collecting and using the data.
“I don’t know any other school that goes into the ethical side of it,” he said.
There are also some unusual applications for data-mining techniques that others may be overlooking, Simon said. Researchers could scan historical texts for certain words or phrases to get a better sense of the period. New technology means they can cast a wider net than ever before.
Board of Visitors members were supportive of the idea, but some seemed perplexed as to why it’s such a prominent part of the strategic plan. Administrators presented parts of the strategic plan to board members, and big data took up much of the lecture.
“I feel like we turned into the big data design committee,” said first-year board member Kevin Fay, a sentiment that was echoed in some way or another by several other board members.
But they agreed it’s a good idea and a subject the university ought to get into.
Simon said, speaking after the meeting, that the institute will serve as a model for the kind of “strategic research” – research that serves the public and that the university can excel in – UVa wants to do more of in the future.
If they can work out the kinks, such as finding a way to design courses involving multiple departments, and attract federal funding, administrators will be pushing for more research institutes down the line.
“It’s really the first of its kind here,” Simon said. “The next one will be easier.”
Weather JournalRain is here; watching for snow