Enhanced middleware for collaborative privacy in community based recommendations services

Recommending communities in social networks is the problem of detecting, for each member, its membership to one of more communities of other members, where members in each community share some relevant features which guaranteeing that the community as a whole satisfies some desired properties of similarity. As a result, forming these communities requires the availability of personal data from different participants. This is a requirement not only for these services but also the landscape of the Web 2.0 itself with all its versatile services heavily relies on the disclosure of private user information. As the more service providers collect personal data about their customers, the growing privacy threats pose for their patrons. Addressing end-user concerns privacy-enhancing techniques (PETs) have emerged to enable them to improve the control over their personal data. In this paper, we introduce a collaborative privacy middleware (EMCP) that runs in attendees' mobile phones and allows exchanging of their information in order to facilities recommending and creating communities without disclosing their preferences to other parties. We also provide a scenario for community based recommender service for conferences and experimentation results.