Privacy-Preserving Data Mining of Query Logs with Multiple Log Subtables in Conditional Functional Dependencies

This paper presents the anonymization of query logs using d,ℓ inference models. Proposal ensures the d,ℓ diversity of the users in the query log, while preserving its utility. The recent release of the American Online (AOL) Query Logs deserved high. So proposed d,ℓ diversity is applied to AOL query log files and attacks against Conditional Functional Dependencies (CFDs) as well as web log query attacks. Effectively prevent against an CFDs -based privacy attack. Formalize the CFD-based privacy attack and define the privacy model, (d;ℓ)-inference, to combat the CFDs -based attack. All the methods are novel and represent additional options for web log data anonymization. Provide the evaluation of (d;ℓ)-inference model in real query logs, showing the privacy and utility achieved, as well as providing estimations for the use of such data in data mining processes.