London, Mar 30 (ANI): With more and more people drawing towards social networking sites, it has become very hard to maintain privacy, according to a new study.
The researchers analysed links between users of social sites, and identified many people in supposedly anonymous data sets.Social sites produce the anonymised data, and sell it to marketing firms for generating cash.
The researchers claim that the results clearly indicate that web firms should put in more effort to protect users' privacy.
For the study, computer scientists Arvind Narayanan and Dr Vitaly Shmatikov, from the University of Texas at Austin, developed an algorithm that turned the anonymous data back into names and addresses.
Usually, personally identifiable information, such as names, are removed from the data sets.
And then they are sold to marketing companies or researchers keen to plumb it for useful information.
Earlier, it was believed that removing this data is enough to know that it's impossible to reconstruct the true identities of subjects.
The algorithm analyses not only analyses the immediate friends that members of these sites connect to, but also take into account the relationships between all the members of a social network.
The researchers used Social graphs from Twitter, Flickr and Live Journal for their study, reports the BBC.
It was found that one third of people-with their profiles on both Flickr and Twitter-can be identified from the completely anonymous Twitter graph, despite the fact that the overlapping of members between the two services is thought to be about 15 percent.
The researchers suggested that as social network sites become more heavily used, people would find it increasingly difficult to maintain a veil of anonymity.
The pair will present a paper about their work to the IEEE Symposium on Security and Privacy taking place in California from 17-20 May. (ANI)