A Homomorphic Encryption Approach to Implementing Two-Party Privacy Preserving Data Mining

Abstract

The problem of privacy-preserving data mining has become more important in recent years because of the increasing ability to store personal data about users, and the increasing sophistication of data mining algorithm to leverage this information. This paper is aimed at developing a protocol that will address the issue of privacy in data mining tasks in two party scenarios. We have proposed a framework that uses the homomorphic encryption to add security so that any data mining technique does not lose its valuable data. With the aid of this approach, confidentiality at both parties end is achieved. This model gives valid data mining results for analysis purpose but the actual or true data is not revealed. We discussed implementation evaluation based on metrics proposed by [1], for the framework and algorithm proposed. Tools used include PHP programming language for simulation, MYSQL for database and Visual Paradigm for the modeling analysis. Secondary sources such as academic literature and technical literature from the internet were studied for further classification of processes and techniques.