Privacy Preserving Association Rule Mining based on the Intersection Lattice and Impact Factor of Items
B. Janakiramaiah, RamaMohan Reddy A, Gunjan Kalyani, Hs Mic · 2013
Association Rules revealed by association rule mining may contain some sensitive rules, which may cause prospective threats towards privacy and protection. A number of researchers in this area have recently made efforts to preserve privacy for sensitive association rules in transactional databases. In this paper, we put forward a heuristic based association rule hiding algorithm to get rid of the sensitive knowledge from the released database based on the intersection lattice of an item. The projected algorithm specifies the victim item based on the concept of impact factor of an item in the sensitive rule on the non sensitive frequent item sets. The impact factor of an item in the sensitive association rule is equal to the number of non sensitive frequent item sets that are affected by removing that item from the required number of transactions. Lower the impact factor of an item, lower is its effect on the non sensitive frequent item sets. Proposed algorithm exhibits the concept of intersection lattice and impact factor to conceal several rules by modifying less significant number transactions. As modifications are fewer, data excellence is very less exaggerated.