Vladimir Braverman likens the algorithms that he and his team are inventing to deal with massive amounts of streaming data to the hammers and picks that miners use to remove precious metals from the Earth.
“In order for us to extract patterns and valuable information that are hiding in all that data, we need to build strong new hammers,” says the computer scientist, a co-principal investigator on a three-year, $1 million National Science Foundation grant aimed at developing algorithms that will allow researchers to better extract information from massive streams of data.
These tools have the potential to help astronomers create more accurate cosmological simulations, to allow neuroscientists to better analyze abnormal connectivity patterns in brain graphs, and even to enable researchers to identify meaningful patterns or bursts of significant activity within massive text streams like Twitter.
“Our goal is to create simple tools and methods that allow data analysis to be accessible to almost anyone,” Braverman explains.
The team includes fellow computer scientists Randal Burns, Benjamin Van Durme, and Bloomberg Distinguished Professor Alexander Szalay, as well as Tamas Budavari of the Department of Applied Mathematics and Statistics.