A Greedy-based Approach for Hiding Sensitive Itemsets by Transaction Insertion
Jerry Chun‐Wei Lin, Tzung‐Pei Hong, Chia-Ching Chang, Shyue-Liang Wang · J. Inf. Hiding Multim. Signal Process. · 2013
Data mining technology is designed to derive useful knowledge from large database, which is used to aid decision making. The process of data collection and dis- semination may, however, causes privacy concerns. Sensitive or personal information and knowledge of individuals, industries and organizations must be kept private before they are publicly shared or published. Thus, privacy-preserving data mining (PPDM) has become an important issue to efficiently hide sensitive information. In this paper, a greedy-based approach is proposed to hide sensitive itemsets by transaction insertion. The proposed approach first computes the maximal number of transactions to be inserted into the original database for totally hiding sensitive itemsets. The fake items of the transac- tions to be inserted are thus designed by the statistical approach, which can greatly reduce side effects in PPDM. Experiments are also conducted to evaluate the performance of the proposed approach.