Hiding association rules based on relative-non-sensitive frequent itemsets

Abstract

Association rules hiding algorithms often sanitize transactional databases for protecting sensitive information. Data modification is one of the most important sanitation approaches. However, the exist modification methods either focus on hiding sensitive rules only, or take measures to reduce the impact on non-sensitive rules from the whole database while hiding sensitive rules. In this paper, we propose a new algorithm which hides sensitive rules from the side of non-sensitive rules. It classifies the sensitive transactions by their degree of conflict. For the special group of transactions, a victim-item must satisfy: 1, in the sensitive rules; 2, not in the non-sensitive rules. Our algorithm selects different victim-items in different transactions that contain the same rule, which makes sure that removing the victim-items in the special group of transactions has no influence to non-sensitive rules. The experimental results show that our algorithm for sanitizing transactional database can achieve better results compared with others algorithms such as Naïve, MinFIA, MaxFIA and IGA. In particular, our algorithm has the least impact on non-sensitive rules.