We investigate the effects of different privacy enhancing technologies in content-based recommendation systems.
We study the interplay between the degree of privacy and the potential degradation of the quality of the recommendation.
We evaluate three different tag forgery strategies: optimised tag forgery, uniform tag forgery and TrackMeNot.
We carry out an experimental evaluation on a real dataset extracted from Delicious.
Recommended citation: S. Puglisi, J. Parra-Arnau, J. Forné, D. Rebollo-Monedero. (2015). “On Content-Based Recommendation and User Privacy in Social-Tagging Systems” Computer Standards & Interfaces. 41 (17-27).