Privacy-Preserving Outsourced Mining of D-Eclat Association Rules on Vertically Partitioned Databases

Abstract

Association rule mining and frequent item set mining are extensively studied information analysis techniques for a number of programs. In this paper, we have highlighted privacy preserving mining on vertically partitioned databases. In such a state of affairs, information owners desire to research the association policies or frequent item sets from a collective dataset, and reveal as meager information concerning their delicate information as conceivable to different statistics owners and outsiders. To ensure information security, we outline an efficient homomorphic encryption scheme and a comfy assessment scheme. We then recommend a cloud-aided frequent item set mining answer which develops an association rule mining solution. Our answers are framed for outsourced databases that permit multiple facts owners to effectively proportion their data securely without compromising on facts privacy. Our answers release less information with respect to the information than most existing arrangements. Considering each information and computation work is outsourced to the cloud servers, the useful resource intake at the information owner may be very low. This paper makes use of D-Eclat algorithm for association rule mining over vertically partitioned databases which is more efficient than the Eclat algorithm.