EFPA: Efficient and flexible privacy-preserving mining of association rule in cloud

Abstract

With the explosive growth of data and the advance of cloud computing, data mining technology has attracted considerable interest recently. However, the flourish of data mining technology still faces many challenges in big data era, and one of the main security issues is to prevent privacy disclosure when running data mining in cloud. In this paper, we propose an efficient and flexible protocol, called EFPA, for privacy-preserving association rule mining in cloud. With the protocol, plenty of participants can provide their data and mine the association rules in cloud together without privacy leakage. Detailed security analysis shows that the proposed EFPA protocol can achieve privacy-preserving mining of association rules in cloud. In addition, performance evaluations via extensive simulations also demonstrate the EFPA's effectiveness in term of low computational costs.