Privacy in Search Logs

By M. Götz, Ashwin Machanavajjhala, Guozhang Wang, Xiaokui Xiao, Johannes Gehrke. 2009

37 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Boosting and Differential Privacy (2010, background): “We note that there is another notion, known as (ε, δ)-probabilistic differential privacy [18], [14], which lies strictly between ε-differential privacy and (ε, δ)-differential privacy.”
Privacy via the Johnson-Lindenstrauss Transform (2012, methodology): “On the other hand, approaches to data sharing such as the recently proposed provably private methods for search query and click data release [16,20] accomplish it at the price of giving up on all user-level information.”
Differential privacy via wavelet transforms (2009, methodology): “In particular, the solutions by Korolova et al. [14] and G¨otz et al. [15] are developed for releasing query and click histograms from search logs.”
Publishing Search Logs—A Comparative Study of Privacy Guarantees (2009, background): “[17] propose the same basic algorithm that we propose in [10] and review in Section 4.”
Discovering frequent patterns in sensitive data (2010, methodology): “(cid:0) (cid:1) In the experimental settings we consider, the value of (cid:0) (cid:1) is far higher than the maximum value ω (i.e., the number of itemsets whose frequencies change by changing one transaction in the data set) used by [15] and [19].”
On sampling, anonymization, and differential privacy or, k-anonymization meets differential privacy (2011, methodology): “et al. [18] considered publishing search queries and clicks that achieves (ǫ,δ)-differential privacy. A similar approach for releasing query logs with differential privacy was proposed by Gotz et al. [15]. These approaches essentially apply the output perturbation technique in differential privacy t”
A Differentially Private Stochastic Gradient Descent Algorithm for Multiparty Classification (2012, result): “Before this we define the notion of ( , δ)-differential privacy [10] which will be used in our results.”