Skip to main content

Mobile Users Effectively Control Privacy in a Friend-Finder Application

Researchers: Norman Sadeh, Lorrie Cranor, Jason Hong

Research Area: Mobility | Privacy Protection

Cross Cutting Thrusts: Usable Privacy and Security

Abstract

The objective of this project is to develop and validate technologies that empower users to more effectively and more efficiently control their privacy policies in mobile and social networking applications. Specifically, as part of this project, we will:

  • Deploy our PeopleFinder application and make it available to the Carnegie Mellon campus community. This will include scaling up our technology and conducting the largest evaluation so far of such a system.
  • Systematically study key tradeoffs between policy accuracy, user burden and overall sense of control, as we deploy policy authoring tools implementing novel dialogue, explanation and machine learning technologies. We propose to focus on the development and evaluation of a new family of user-controllable policy learning techniques and expandable grids policy authoring technology.
  • Conduct the first systemic study aimed at understanding how people's privacy policies evolve over long periods of time.

This research is particularly timely as it addresses privacy issues at the core of new mobile and social networking applications. While a number of these applications have encountered some initial success, there is a growing demand among the user community for higher levels of control over how one's information is being shared by others. This project is uniquely positioned to produce technologies that respond to this issue. By scaling up to thousands of users, we will make it possible to develop and validate technologies in a realistic setting, generate significant publicity and attract the interest of major players in the mobile and social networking space.