Big Data: Is Anonymity a Possibility?

We live in the era of Big Data: one in which we create 2.5 quintillion bytes of data a day. In fact, we create so much data that 90% of the world’s data today has been created in the last two years alone, according to IBM. The typical American office worker, for example, produces about 5,000 megabytes a day, including Word files, downloaded movies and email, as that information is moved along mobile networks or across the Internet. The explosive growth of mobile media – and the massive amounts of location-based data it creates each day – has contributed to this phenomenon.

 

The typical American office worker produces 1.8 million megabytes of data each year.

 

Today, almost any type of data can be used to identify the person who created it ~ pretty much like a fingerprint. In effect,   the more data we generate, the less private it can be:  the greater the amount of personal data that’s available, the more informative the data gets. Eventually, data mining may render  anonymity of any kind may be “algorithmically impossible,” says Arvind Narayanan of Princeton University.

Mashable explains how  information we once thought of as personal data, such as our name and address, are already bought and sold by data brokers. Acxiom, for example, holds an average of 1,500 pieces of information on over 500 million consumers around the world, such as  age, education, and zip code, car’s make and model, your income and investments. 

Based on this information, each person is placed in one of 70 different “PersonicX” clusters, which are “summarized indicators of lifestyle, interests and activities.” While these data sets are often depicted as having been “anonymized,” the more data they incorporate, the less likely this is to be actually true.

 

Case in Point:Location Data

Mobile-phone companies record users’ locations, strip  the phone numbers, and then sell the aggregate data sets. However, even when such location data is anonymous, it can take  just four different data points about a phone’s position to link that phone to a unique person.

MIT researchers Yves-Alexandre de Montjoye and César A. Hidalgo of MIT have shown that even when such location data is anonymous, just four different data points about a phone’s position can usually link the phone to a unique person. In fact, with enough data, it’s may be even possible to discover information about a person’s future.

Source: Mashable

TED

The Next Web.