How Dangerous Are Your Smartphones? App Usage Recommendation with Privacy Preserving

Abstract

With the rapid proliferation of mobile devices, explosive mobile applications (apps) are developed in the past few years. However, the functions of mobile apps are varied and the designs of them are not well understood by end users, especially the activities and functions related to user privacy. Therefore, understanding how much danger of mobile apps with respect to privacy violation to mobile users is becomes a critical issue when people use mobile devices. In this paper, we evaluate the mobile app privacy violation of mobile users by computing the danger coefficient. In order to help people reduce the privacy leakage, we combine both the user preference to mobile apps and the privacy risk of apps and propose a mobile app usage recommendation method named AppURank to recommend the secure apps with the same function as the “dangerous” one for people use. The evaluation results show that our recommendation can reduce the privacy leakage by 50%.