Privacy-based recommendation mechanism in mobile participatory sensing systems using crowdsourced users preferences

Participatory sensing has been pioneered as a novel sensing pattern to collect and interpret information from the environment using ubiquitous and ever-more-capable mobile devices. One of the main obstacles for long-term participation in such systems are users privacy concerns. Due to the nature of these systems, users have to agree to provide some personal information, which may lead to privacy disclosure. This risk will dampen users enthusiasm to participate in sensing activities, and diminish the advantage of participatory sensing accordingly. To mitigate the privacy risk, our basic is to make recommendations to users about what data can be provided based on their privacy preferences when they are in the mobile participatory sensing systems. We propose PriRe, a privacy-based recommendation mechanism. More specifically, PriRe first measures privacy risks based on user preferences towards data sharing in participatory sensing systems and then make the recommendation according to the measurement. Further, we implemented and deployed PriRe in the real world as a user study for evaluation. The study shows that PriRe can measure the users privacy accurate and provide effective recommendations to users for data sharing in mobile participatory sensing systems. It is also accepted by the users as a trustworthy tool.