Search engine firms collect the "database of intentions," the histories of their users' search queries. These search logs are a gold mine for researchers. Search engine firms, but, are wary of publishing search logs so as not to disclose sensitive data. In this paper we have a tendency to analyze algorithms for publishing frequent keywords, queries and clicks of a pursuit log. We tend to 1st show how methods that achieve variants of k-anonymity are at risk of active attacks. We tend to then demonstrate that the stronger guarantee ensured by epsilon-differential privacy sadly will not provide any utility for this downside. We have a tendency to then propose a novel algorithm ZEALOUS and show how to line its parameters to realize (epsilon, delta)-probabilistic privacy. We conjointly distinction our analysis of ZEALOUS with an analysis by Korolova et al. that achieves (epsilon', delta')-indistinguishability. Our paper concludes with a large experimental study using real applications where we tend to compare ZEALOUS and former work that achieves k-anonymity in search log publishing. Our results show that ZEALOUS yields comparable utility to k-anonymity while at the identical time achieving a lot of stronger privacy guarantees.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here