Title
A semantic-preserving differentially private method for releasing query logs.
Abstract
•We discuss the challenges and particularities of privacy-preserving releases of query logs.•We propose a differentially private method that protects query logs from a semantic perspective.•We propose several semantic criteria to probabilistically replace queries while fulfilling differential privacy.•We provide a mechanism to balance the protection and the utility preservation of query logs.•Empirical results show that our method produces protected logs that are more useful for user profiling than related works.
Year
DOI
Venue
2018
10.1016/j.ins.2018.05.046
Information Sciences
Keywords
Field
DocType
Differential privacy,Query logs,Data utility,User profiling
Joins,Information retrieval,Curse of dimensionality,Personally identifiable information,Artificial intelligence,Data Protection Act 1998,Machine learning,Limiting,Mathematics,Semantics,Statistical disclosure control
Journal
Volume
ISSN
Citations 
460
0020-0255
0
PageRank 
References 
Authors
0.34
29
5
Name
Order
Citations
PageRank
David Sánchez169033.01
Montserrat Batet289937.20
Alexandre Viejo335225.61
Mercedes Rodriguez-Garcia482.53
Jordi Castellà-Roca501.35