Deep Privacy Concerns

Abstract

Online video-streaming company Netflix once released hundreds of millions of items of “anonymously processed” data on moving ratings, preserving only each user’s rating of the movie and the timestamp of when they rated it. It hoped to find a better movie recommendation algorithm through the form of a contest. However, in 2009, two researchers at the University of Texas, by comparing these anonymous figures with the open IMDB database, were able to successfully match anonymous data to specific users, and Netflix ultimately had to cancel what was originally planned to be an annual competition. Netflix’s case shows that with big data analytics people’s secrets cannot be hidden anywhere. So-called privacy protections are nothing more than “the emperor’s new clothes.”