Data management professionals working in federal govt. Data architects, modelers, analysts, etc. Includes master data management, metadata, data quality ...use your imagination.
Microsoft and Database Privacy
December 6, 2012 at 11:23 am #174192
from Microsoft Research
The problem of statistical disclosure control—revealing accurate statistics about a population while preserving the privacy of individuals—has a venerable history. An extensive literature spans multiple disciplines: statistics, theoretical computer science, security, and databases. Nevertheless, despite this extensive literature, «privacy breaches» are common, both in the literature and in practice, even when security and data integrity are not compromised.
This project revisits private data analysis from the perspective of modern cryptography. We address many previous difficulties by obtaining a strong, yet realizable, definition of privacy. Intuitively, differential privacy ensures that the system behaves the essentially same way, independent of whether any individual, or small group of individuals, opts in to or opts out of the database. More precisely, for every possible output of the system, the probability of this output is almost unchanged by the addition or removal of any individual, where the probabilities are taken over the coin flips of the mechanism (and not the data set). Moreover, this holds even in the face of arbitrary existing or future knowledge available to a «privacy adversary,» completely solving the problem of database linkage attacks.
Databases can serve many social goals, such as fair allocation of resources, and identifying genetic markers for disease. Better participation means better information, and the «in vs out» aspect of differential privacy encourages participation.
December 6, 2012 at 11:26 am #174196
Title: Differential Privacy for Everyone
Big data technologies offer promise and bring potential concerns. Society can only reap the full benefits offered by the data age if the privacy of individuals is protected at the same time. Microsoft believes that in order for society to reap the full benefits offered by the data age and the creative efforts of researchers and developers, without significantly eroding individual privacy, we will have address a variety of different needs and requirements. For some use cases, leveraging new and innovative privacy-protecting technologies like Differential Privacy will help meet those requirements.
Download: PDF file
December 6, 2012 at 11:38 am #174194
Microsoft Blog Commentary on Differential Privacy:
The Promise of Differential Privacy
At Microsoft, we have some of the world’s top privacy researchers working on a wide variety of interesting challenges. We strive to translate this research into new privacy-enhancing technologies.
Today, we’re releasing a new whitepaper on Microsoft’s research in Differential Privacy written by Javier Salido on my team. To help set the stage, I’d like to provide some background on this timely topic.
Over the past few years, research has shown that ensuring the privacy of individuals in databases can be extremely difficult even after personally identifiable information (e.g., names, addresses and Social Security numbers) has been removed from these databases. According to researchers, this is because it is often possible, with enough effort, to correlate databases using information that is traditionally not considered identifiable. If any one of the correlated databases contains information that can be linked back to an individual, then information in the others may be link-able as well.
Differential Privacy (DP) offers a mathematical definition of privacy and a research technology that satisfies the definition. The technology helps address re-identification and other privacy risks as information is gleaned from a given database. Differential Privacy does this by adding noise to the otherwise correct results of database queries. The noise helps prevent results of the queries from being linked to other data that could later be used to identify individuals. Differential privacy is not a silver bullet and needs to be matched with policy protections, such as commitments not to release the contents of the underlying database, to reduce risk.
You must be logged in to reply to this topic.