A Privacy Providing Context-based Trust Model for OSN and its Relation to GDPR

Abstract Users’ privacy has become a key objective inthe operation of Online Social Networks (OSN), whichinvolves major concerns regarding personal data exposure and processing. It became a demand with the raiseof legislative initiatives such as the General Data Protection Regulation (GDPR) in the EU that affects mostof the commercial companies, government institutions,and other sectors that process personal data. While thesubject of ensuring privacy in OSN is widely addressedin current research, we focus on challenges related toseveral principles concerning the right to the protectionof personal data, as defined by the GDPR. In previousresearch we have devised a comprehensive Trust-basedModel for Social Networks, that uses Trust, Access Control and Flow Control, to enhance users’ control overtheir data. Users of social networks share data of various contexts, and for a specific user, some contexts aremore sensitive than others. In this paper we use contextdependent trust to build a refined model for controllingpersonal content that aligns with the users’ preferences.We propose a context-based trust model that involvesuser actions, OSN features and OSN activities. It provides a means for revealing the sources of sensitive datadissemination and prevents them from happening. Wevalidate this model by analyzing trust using sentimentanalysis of posts in a real network. The context-basedmodel we present in this paper, addresses the need toestablish an infrastructure for the enforcement of dataprotection rules, concerning users’ privacy in OSN, suchas data minimization and the right-to-be-forgotten. Wealso propose a solution for Digital Rights Management(DRM) to control and manage access to copyrightedmaterial.