A privacy-aware model to process data from location-based social media
Marc Löchner, Alexander Dunkel, Dirk Burghardt · KOPS (University of Konstanz) · 2018
Many social media services offer their users to add location data to their posts. Since this data is usually publicly available, it can be used to create thematic maps based on the topical information, e.g. derived from hashtags attached to posts. However, users might not be aware, that their publications can be used for other purposes by third parties. In certain situations it can be compromising the user’s privacy. We introduce a conceptual model to help people who create those maps to preserve privacy of social media users. Therefore we analyze the data in a set of four facets. For each facet, we eliminate precise data by deriving multiple abstraction layers from it. Using these layers, we are able to quantitatively describe different levels of privacy. We further describe an example application to show use cases for the abstraction model on the same data in two contrary scenarios.