Research Spin-offs from Commercial Data Mining

By administrator | May 17, 2006

Edmonton Journal technology columnist Steve Makris wrote about recent developments in data mining that could have potential implications for providing access to confidential social survey data (see “Hot Data,” Edmonton Journal, Business, G1, May 17, 2006.) Makris mentioned some major companies raiding each other’s data mining experts, such as Google hiring Kai-Fu Lee away from Microsoft only to see Microsoft swoop up Rakesh Agrawal from IBM. Can these companies capitalize on the massive data accumulated by private companies and governments through data mining technology?

[Makris] chatted with Agrawal in his tiny and very busy office at IBM’s San Jose Almaden Research Centre during a Silicon Valley tour last year. He was taking data mining to the next level: Sovereign Information Integration … a cryptographic-like way of allowing sharing of data by controlling users’ access privileges. This would let even the fiercest competitors to data mine common, neutral data without disclosing sensitive secrets and possibly lead to beneficial partnerships. “Today you can have both security and privacy and still share information,” said Agrawal.

While short on actual detail, the news of commercial investment in technology to share private information is worth monitoring for the social data research community.

  • Contributed by Chuck Humphrey