Privacy Preservation in Data Mining Using Elliptical Curve Cryptography

There are many distributed and centralized data mining techniques often used for various applications. Privacy and security issues of these techniques are recently investigated with a conclusion that they reveal information or data to each other clients involved to find global valid results. But because of privacy issues, involving clients do not want to share such type of data. Recently many cryptography algorithms have been found to address privacy problems in distributed and centralized data mining. In this thesis, we propose an elliptic curve cryptography based algorithm to mine privacy-preserving association rules on horizontally partitioned data. Elliptic Curve Integration Encryption Scheme is used for security of data and Elliptic Curve Digital Signature Algorithm for authentication. Moreover, we have also considered unsecured communication channels in distributed environment. Proposed algorithm provides privacy and security against involving clients and other clients (adversaries) who can reveal information by reading unsecured channel between involving clients. Finally, we analyse the privacy and security provided by proposed algorithm.