Abstract |
Online social networks have emerged with the development of Web 2.0 and evolved to the hugely popular services that billions of people currently use on a regular basis. They currently provide a wide range of functionalities, from content sharing to online gaming, but the main reason behind their success is that they provide a solution to one of the most basic human needs, the need for communication. Without doubt, the proliferation of online applications and services, and in particular the emergence and rapid growth of online social networks, has radically transformed the online activities and behavior of users, who currently spend a quite large portion of their time on these services. In an effort to utilize the capabilities offered by the Internet at the fullest, people have started becoming more and more reliant on the Web and the services it provides, for completing even the most simple daily tasks, carrying out their work, spending their free time, as well as communicating and interacting with each other. However, OSNs have also introduced several privacy issues in people's life. The implications of these privacy issues become alarming when considering the volume of user personal information and data published online, and also considering the popularity and scale of adoption of these services. During the last years we observe a continuous effort by the research community, the media and data protection agencies towards exploring the problem of privacy in OSNs, making users more aware about the implications of disclosing sensitive information, and proposing solutions to mitigate these threats. Although the demand for privacy pushed OSNs towards implementing more privacy-oriented policies and mechanisms, the problem of privacy has not been entirely solved yet. In the dissertation we explore the privacy problems that arise from: (i) OSNs' poorlyconceived data management and sharing policies, and (ii) the mechanisms of OSNs for the resolution of conflicts of interest and for collective resource management. Specifically, we investigate the risk users face due to the current policies of Twitter for the collection, management and sharing of users' precise geo-location information through its publicly available API. Additionally, we investigate the problem of conflicts of interest and users' conflicting privacy settings for collective resources, and we design two fine-grained access control mechanisms that enable the users associated with a collective resource to control the exposure of sensitive information of their interest, while ensuring that the privacy preferences of each user are not affected by those of other users associated with the resource.
|