Patronus: Augmented Privacy Protection for Resource Publication in Online Social Networks

On-line social networks are playing an important role in our daily lives in the modern society. It is becoming much more frequent and convenient for individuals to communicate with others by publishing data such as photos and videos over on-line social networks. As plenty of personal information is usually contained in these data, privacy leakage has become a more serious problem. However, current privacy protection mechanism, which is built on the group-based access control model (GBAC), are typically owner-centric. It means that the privacy requirements of data involvers, whose privacy information is also contained in the corresponding resource, are typically overlooked and ignored. In order to provide augmented privacy protection for social data publication, this paper proposes a new data access model named Patronus to provide better privacy protection during social data publication. With Patronus, the privacy requirements of the data owner and its involvers are both taken into considerations. In order to automatically recognize the corresponding involvers and enforce their privacy polices, Patronus provides a simplistic specification based on the format of ``when-where-who-what´´ to describe the features of a resource and the privacy preferences of an individual user. We implemented a prototype of Patronus for photo sharing on Android, and demonstrated its feasibility and effectiveness with several case studies.