An interesting new direction for data mining research is the development of techniques that incorporate privacy concerns for association rules. In this work, we present a framework for mining association rules from various transactions. These transactions mainly consisting of categorical items, where the data has to preserve privacy of individual transactions. By using uniform randomization, it is feasible to recover association rules, but these rules are in turn be exploited to find privacy breaches. Hence, in this work we clearly analyze the nature of privacy breaches and propose a new class of randomization operators that are much more effective than uniform randomization which was proposed previously. Here we also derive formulae for an unbiased support estimator, which allows us to recover item set supports from randomization data sets. Here we also show how the above derived formulae will be incorporated into mining algorithms. Finally; we provide experimental results that validate the proposed algorithm by applying it to real data sets.

Privacy Preserving Mining uniform randomization