Completely hide sensitive association rules using EMO by deleting transactions
Peng Cheng, Jeng‐Shyang Pan · 2014
Data mining techniques enable efficient extraction of useful knowledge from a large data repository. However, it also can disclose sensitive information if used inappropriately. A feasible way to address this problem is to sanitize the database to conceal sensitive information. In this paper, we focus on privacy preserving in association rule mining. In light of the tradeoff between hiding sensitive rules and disclosing non-sensitive ones during the hiding process, a novel association rule hiding approach is proposed based on evolutionary multi-objective optimization (EMO). It modifies the original database by deleting identified transactions/tuples to hide sensitive rules. Experiment results are reported to show the effectiveness of the proposed approach.