Analysis and Design of Privacy-Enhancing Information Sharing Systems

In the context of digital environment, Online Social Networks (OSNs) are very popular among Internet users: a 2010 survey reported nearly half of the online users are members of an OSN, with Facebook reaches more than 1 billion monthly active users and Twitter 140 million. About 35% of active users spend more time online than face-to-face and 33% of them were more likely to speak to someone online than offline. OSNs havealtered the social ecosystem in a remarkable way by providing users’ with a plethora of easy-to-use tools, offering highly efficient means for establishing and maintaining social connections. Users are allowed to share information through text, pictures and videos, and build communitiesaround shared interests. According to Facebook, a user with 130 friendsand a membership in 80 groups and events submits, on average, 90 piecesof data per month. The vast increase of users, sharing options and tools, leads to the dissemination of large amounts of information in OSNs; users disclose information related to them or to their friends, as they share content. For instance, on Facebook users share personal opinions, pictures and submit their current activities. This information is accessible by a diverse set of viewers, such as online friends, service providers and third party organizations. Henceforth, this has prompted serious privacy concerns among the users; as reported by the media and the research community. These concerns increased after the US government mass surveillance program (including PRISM) became known to the public in 2013. Hence, we identified two main privacy and trust open research questions in online communities: (i) How much of the contributed informationcan be implicitly exposed without the knowledge of the actual user ? (ii) How can we protect users from this leak ? Motivated by the above concerns, this PhD will explore a number of potential ways to enhance trust and privacy in online communities. The goal is to design user-centered model to preserve privacy and trust for online communities.status: accepte