1. What are the contributions in "Evaluating tag-based preference obfuscation systems" ?
This paper considers an easily implementable class of obfuscation strategies as a means to mitigate these risks, and examines its privacy/utility tradeoff.. The authors conducted experiments with different simulated behaviours and using two preference populations, namely a population of maximally diverse preferences and one consisting of the movie preferences of some Netflix users.
read more
2. What have the authors stated for future works in "Evaluating tag-based preference obfuscation systems" ?
Future work items include the development of adversaries that also observe differing favourable tags and have various degrees of background knowledge about user behaviour, and examining the case where obfuscation parameters vary from user to user.. The authors also intend to extend their simulation framework for measuring unlinkability in other settings such as electronic cash and, more generally, attribute exchange.
read more
![Fig. 7. Advantages a(GRA, β) and a(SAA, β) as a function of β, for all β ∈ [0.4, 1] and for all n ∈ {30, 31, . . . , 75}. Each value of n results in a different curve (see text).](/figures/fig-7-advantages-a-gra-b-and-a-saa-b-as-a-function-of-b-for-1fx7cwn0.webp)