P3ARM-t: Privacy-Preserving Protocol for Association Rule Mining with t Collusion Resistance

摘要

—The ability to mine large volumes of distributeddatasets enables more precise decision making. However,privacy concerns should be carefully addressed when miningdatasets distributed over autonomous sites. We proposea new cryptography-based Privacy-Preserving Protocolfor Association Rule Mining with t collusion resistance(P3ARM-t), where t is the threshold of number of colludingsites. P3ARM-t is based on a distributed implementationof the Apriori algorithm. The key idea is to arbitraryassign polling sites to collect itemsets’ supports in encryptedforms using homomorphic encryption techniques. Pollingsites are randomly assigned and are different for consecutiverounds of the protocol to reduce the potential for collusion.Our performance analysis shows that P3ARM-t significantlyoutperforms a leading existing protocol. Moreover, P3ARMtis scalable in the number of sites and the volume of data.The protocol also decreases the potential for collusion forup to t colluding sites.