A conceptual framework for privacy preserving of association rule mining in e-commerce

This paper proposes a conceptual framework for privacy preserving of association rule mining in e-commerce and a novel heuristic approach for association rule hiding based on intersection lattice theory. Our approach relies on characteristics of intersection lattice and frequent itemsets to specify exactly victim items in such a way that modifying them leads to reduction of confidence of the sensitive association rules but less impact on remains association rules. The experimental results indicate effectiveness of the proposed algorithm in comparison with previous work.