In this day and age with the prevalence of smartphones, networking has evolved in an intricate and complex way. With the help of a technology-driven society, the term "social networking" was created and came to mean using media platforms such as Myspace, Facebook, and Twitter to connect and interact with friends, family, or even complete strangers. Websites are created and put online each day, with many of them possessing hidden threats that the average person does not think about. A key feature that was created for vast amount of utility was the use of location-based services, where many websites inform their users that the website will be using the users' locations to enhance the functionality. However, still far too many websites do not inform their users that they may be tracked, or to what degree. In a similar juxtaposed scenario, the evolution of these social networks has allowed countless people to share photos with others online. While this seems harmless at face-value, there may be times in which people share photos of friends or other non-consenting individuals who do not want that picture viewable to anyone at the photo owner's control. There exists a lack of privacy controls for users to precisely de fine how they wish websites to use their location information, and for how others may share images of them online. This dissertation introduces two models that help mitigate these privacy concerns for social network users. MoveWithMe is an Android and iOS application which creates decoys that move locations along with the user in a consistent and semantically secure way. REMIND is the second model that performs rich probability calculations to determine which friends in a social network may pose a risk for privacy breaches when sharing images. Both models have undergone extensive testing to demonstrate their effectiveness and efficiency.
Privacy vs. Reward in Indoor Location-Based Services