Privacy Preserving Association Rule Mining Using Binary Encoded NSGA-II

When people utilize data mining techniques to discover useful knowledge behind a large database; they also have the requirement to preserve some information so as not to be mined out, such as sensitive or private association rules, classification tree and the like. 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 within the side effects accompanying the hiding process, we tackle this problem from a point view of multi-objective optimization. A novel association rule hiding approach was proposed based on evolutionary multi-objective optimization (EMO) algorithm. The binary encoding scheme was adopted in the EMO algorithm. Three side effects, including sensitive rules not hidden, nonsensitive lost rules and spurious rules were formulated as objectives to be minimized. The NSGA II algorithm, a well established EMO algorithm, was utilized to find a suitable subset of transactions to modify by removing items so that the three side effects are minimized. Experiment results were reported to show the effectiveness of the proposed approach.