The Computer Science and Artificial Intelligence Laboratory (CSAIL) at the Massachusetts Institute of Technology (MIT) has announced a new program for exploring and improving the use of Big Data.
Bigdata@CSAIL will bring together representatives from academia, industry and government to develop and improve methods of collecting, processing, analyzing, storing and sharing massive datasets made possible by Big Data with the goal of making them more useful for society,
to a CSAIL press release.
“With the right tools, we can begin to make sense of the data and use it to solve any number of pressing societal problems—but our existing tools are outdated and rooted in computer systems and technologies developed in the 1970s,” said MIT Department of Electrical Engineering and Computer Science Associate Professor Sam Madden.
Hardware and software developers, theoretical computer scientists and computer security experts will help develop infrastructures and algorithms to make more efficient and accurate use of Big Data. These researchers will work in areas such as finance, healthcare, social media and security.
“We hope to develop more sophisticated tools for in-depth processing of medical information, which could lead to more accurate diagnostic techniques and better treatment methods for patients,” Madden said, adding, “We also want to secure the ever-expanding datasets of medical, financial and personal information.”
CSAIL Director Daniela Rus noted the initiative’s “approach to Big Data is unique in that we will be building, from the ground up, new methods for dealing with the data deluge, and then applying our techniques to specific research areas.”
The growing use of Big Data has also given rise to privacy and data protection
among privacy advocates and lawmakers.
A recent
conducted by the Pew Research Center and Elon University revealed that 39 percent of respondents said Big Data will likely be “a big negative for society in all respects.”
Some privacy advocates worry that, with increased collection of personal information, results from profiling and algorithms “might draw the wrong conclusion about who someone is, how she might behave in the future and how to apply the correlations that will emerge in the data analysis.” Other respondents expressed concern about government surveillance and market manipulation.
Conversely, 53 percent of respondents believe Big Data “could enhance productivity, improve organizational transparency and expand the frontier of the ‘knowable future.”
Solutions to protecting against the manipulation of personal data may not yet be apparent, but researchers at CSAIL are looking into data security solutions.
“We’ve been thinking about how we might help address some of these challenges,” says Nickolai Zeldovich. “One concrete problem that we’ve been thinking about recently is how can we process large amounts of sensitive data while minimizing the risk of that data being stolen or misused—perhaps by an attacker that breaks into a server, or a curious administrator of a cloud platform that might be hosting the data.”
Zeldovich, also an associate professor in MIT’s Electrical Engineering and Computer Science Department, says one solution being explored is encryption. “Only encrypted copies of data are stored on the server,” he says, adding, “even if an attacker breaks in, they wouldn’t be able to decrypt or misuse the data—while at the same time, the legitimate data owner can use their decryption key to decrypt and obtain results computed on that data.”
The Big Data initiative was announced in May at an MIT event attended by Massachusetts Gov. Deval Patrick and representatives from Intel. In March, the Obama administration also
a $200 million
Simultaneously, Intel announced that it is creating a new Science and Technology Center (ISTC) for Big Data at CSAIL. Intel Chief Technology Officer Justin Rattner said, “CSAIL is one of the top places in the world that brings together people who build computational platforms like databases and networks and people who work on algorithms and machine-learning techniques—with people who have expertise in specific domains such as finance, medicine and security.”
Rattner added, “These are the skills we need for taming Big Data.”
Madden and MIT Adjunct Professor Michael Stonebraker will lead the ISTC for Big Data, which will receive $2.5 million per year for at least the next five years. The initiative will also partner with businesses such as AIG, EMC, SAP and Thompson Reuters.
