Your browser does not support JavaScript!

Home    Detection, measurement and defense against third - party trackers on mobile devices : mobile websites vs mobile apps  

Results - Details

Add to Basket
[Add to Basket]
Identifier 000415054
Title Detection, measurement and defense against third - party trackers on mobile devices : mobile websites vs mobile apps
Alternative Title Εντοπισμός , ανάλυση και άμυνα εναντίων βιβλιοθηκών ταυτοποίησης χρηστών σε φορητές συσκευές :στοσελίδες εναντίον εφαρμογών
Author Παπαδόπουλος, Παναγιώτης-Ηλίας Δ.
Thesis advisor Μαρκάτος, Ευάγγελος
Reviewer Ιωαννίδης, Σωτήρης
Παπαδοπούλη, Μαρία
Abstract The vast majority of online services nowadays, provide both a mobile-friendly website and a mobile application to their users. Both of these choices are usually released for free, with their developers mostly gaining revenue by allowing advertisements from ad networks to be embedded into their content. In order to provide more personalized and thus more effective advertisements, ad networks usually deploy pervasive user tracking, raising this way significant privacy concerns. As a consequence, the users do not have to think only their convenience before deciding which choice to use while accessing a service: web or app, but also which one harms their privacy the least. In this master thesis, we aim to respond to this question: which of the two options protects the users' privacy in the best way, websites or apps? To tackle this question, we study a broad range of privacy related leaks comparing several popular apps with their web counterpart. These leaks may contain not only personally identifying information (PII) but also device-specific information, able to crossapplication and cross-site track the user into the network, and allow third parties to link web with app sessions. Finally, we propose an anti-tracking mechanism that enables the users to access an online service through a mobile app without risking their privacy. Our evaluation shows that our approach is able to preserve user privacy by reducing the leaking identifiers of apps by 27.41%, on average, while it introduces a practically negligible latency of less than 1 millisecond per request.
Language English
Issue date 2018-03-23
Collection   School/Department--School of Sciences and Engineering--Department of Computer Science--Post-graduate theses
  Type of Work--Post-graduate theses
Views 328

Digital Documents
No preview available

Download document
View document
Views : 6