06 Dec US Census to Report Less Accurate Data to Reduce Privacy Risks
The US Census Bureau recently announced that it will be enforcing stronger privacy protections by formally adopting a new framework that uses differential privacy to reduce risks.
While the census has always added some uncertainty to its data, “differential privacy,” can determine the amount of randomness that needs to be added to a data set so that an adequate balance between accuracy and privacy can be achieved.
Harvard computer science professor and one of the inventors of differential privacy, Cynthia Dwork, says that the technique it is suited to the statistical analysis of large data sets, which is exactly the challenge faced by the census with its mandate from Title 13 of the U.S. Code to keep each person’s information private, and its responsibility to provide useful data.
Source: NY Times