An Enhanced Scheme for Privacy-Preserving Association Rules Mining on Horizontally Distributed Databases
Xuan Canh Nguyen, Hoài Bắc Lê, Tung Anh Cao · 2012
In this paper, we propose an Enhanced M.Hussein et al.'s Scheme (EMHS) for privacy-preserving association rules mining on horizontally distributed databases. EMHS is based on the M.Hussein et al.'s Scheme (MHS) proposed in 2008 and improves privacy and performance when increasing the number of sites. EMHS uses two servers, Initiator and Combiner, combined with MFI approach to generate candidate set and homomorphic Paillier cryptosystem to compute global supports. Experimental results show that the performance of EMHS is better than MHS in specific databases when increasing the number of sites. A second scheme is also proposed for the other databases.