Contextual localization through network traffic analysis

The rise of location-based services has enabled many opportunities for content service providers to optimize the content delivery based on user´s location. Since sharing precise location remains a major privacy concern among the users, many location-based services rely on contextual location (e.g. residence, cafe etc.) as opposed to acquiring user´s exact physical location. In this paper, we present PACL (Privacy-Aware Contextual Localizer), which can learn user´s contextual location just by passively monitoring user´s network traffic. PACL can discern a set of vital attributes (statistical and application-based) from user´s network traffic, and predict user´s contextual location with a very high accuracy. We design and evaluate PACL using real-world network traces of over 1700 users with over 100 gigabytes of total data. Our results show that PACL (built using decision tree) can predict user´s contextual location with the accuracy of around 87%.