Abstract |
Over the past few years, microblogging social networking services have become
a popular means of information sharing and communication. Although such
services started as a convenient way of sharing small bits of information among
friends, they are currently being used by artists, politicians, news channels, and
information providers to easily communicate with their constituency. Even though
following specific channels on a microblogging service enables users to receive interesting
information in a timely manner, it may raise significant privacy concerns.
For example, the microblogging service is able to observe all the channels that a
particular user follows. This way, it can infer all the subjects a user might be
interested in and generate a detailed profile of this user. This knowledge, being
property of the microblogging service, can be used for a variety of purposes, most
of which are usually beyond the control of the users.
To address these privacy concerns, we propose k-subscription: an obfuscation-based
approach that enables users to follow privacy-sensitive channels, while, at
the same time, making it difficult for the microblogging service to find out their
actual interests. Our method relies on obfuscation: in addition to each privacy-sensitive
channel, users are encouraged to randomly follow k − 1 other channels
they are not interested in. In this way (i) the actual interests of a user are hidden
in random selections, and (ii) each user contributes in hiding the real interests
of other users. Our analysis indicates that k-subscription makes it difficult for
attackers to pinpoint a user’s interests with significant confidence. We show that
this confidence can be made predictably small by slightly adjusting k while adding
a reasonably low overhead on the user’s system.
|