Privacy-Preserving Data Mining based on Integrated Customer Databases from Different Enterprises
Attila Kiss · 2014
The paper is about data mining projects in real applications, where preserving the users’ privacy is important. The aim was to build a secure multiparty computation (SMC) data mining system with SMC data mining algorithm that would be able to solve the task of classification in a horizontally distributed environment with multiple parties trying for a joint data mining project. For solution of this kind of privacy preserving problems we have designed and developed an SMC system with different modules, a client module, a trusted third party and a classification module. We have worked out a new classification method; our k-means based supervised classifier preserves high level anonymity and provides k-anonymity, where k is a user parameter. At the end of the paper a bank example and its results with high accuracy present the efficiency of our system.