Abstract
New trends in pervasive computing allow for hosting user controlled servers for integrating respective user's social spheres. One main feature of such servers is the provision of a single point for managing user's data and resources from various social interaction services (e.g., LinkedIn, Facebook, etc.). A step forward would be to include the collection and integration of different social ...
Abstract
New trends in pervasive computing allow for hosting user controlled servers for integrating respective user's social spheres. One main feature of such servers is the provision of a single point for managing user's data and resources from various social interaction services (e.g., LinkedIn, Facebook, etc.). A step forward would be to include the collection and integration of different social contacts and their live streams (e.g., activity status, live posts, etc.) from these services. Thereby, various privacy issues related to linkability and unwanted information disclosure, could arise. In this paper, we address how we intend to avoid such privacy issues in the EU FP7 funded di.me project when mining users' social spheres from different sources. Our approach uses (1) the detection of semantic equivalence between contacts as portrayed in online profiles and (2) NLP techniques for analysing shared live streams both; for triggering privacy recommendations. The current status is presented and the portability to other environments is shortly discussed.
Altmetric