Privacy Preserving of Vertically Partitioned Data Using Elliptical Curve Cryptography

Abstract

Data mining's privacy problems have become a big problem as a result of financial considerations. This work investigates the topic regarding the issue of privacy-preserving global association rule analysis over several parties in vertically partitioned data to efficiently prevent hidden trends and perform computations across the parties while respecting the privacy of their data. After that, it suggests using Shamir's method of secret sharing to support distributed mining of association rules. We also provide association rule mining method using elliptic curve cryptography. We believe in this unstable distributed system. The algorithmic approach allows for verification between engaged parties while also protecting entrants and involved parties. Finally, we look at the security and privacy that our proposed division offers.