GroupLens: an open architecture for collaborative filtering of netnews
CSCW '94 Proceedings of the 1994 ACM conference on Computer supported cooperative work
Untraceable electronic mail, return addresses, and digital pseudonyms
Communications of the ACM
Introduction to Modern Information Retrieval
Introduction to Modern Information Retrieval
A pseudonymous communications infrastructure for the internet
A pseudonymous communications infrastructure for the internet
Can pseudonymity really guarantee privacy?
SSYM'00 Proceedings of the 9th conference on USENIX Security Symposium - Volume 9
Empirical analysis of predictive algorithms for collaborative filtering
UAI'98 Proceedings of the Fourteenth conference on Uncertainty in artificial intelligence
A Bayesian approach to learning Bayesian networks with local structure
UAI'97 Proceedings of the Thirteenth conference on Uncertainty in artificial intelligence
| Hi-index | 0.00 |
To offer personalized services on the web and on mobile devices, service providers want to have as much information about their users as possible. In the ideal case, the user controls how much of this information is revealed during a transaction. This is a tradeoff between privacy and personalization: if the disclosed profile is too complex, it may become a pseudonym for the user, making it possible to recognize the user at a later time and link different revealed profile parts into one comprehensive profile of the individual. This paper introduces a model for profiles and analyzes it with the methods of probability theory: how much information is revealed and what is the user's probability of staying anonymous. The paper examines how likely it is that a provider can link different disclosed profiles and recommends algorithms to avoid a possible privacy compromise.