Open AccessPrivacy Preserving Decomposable Mining Association Rules on Distributed Data
Author(s) -
Ahmed M. Khedr,
Zaher Al Aghbari,
Ibrahim Kamel
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i3.13.16343
Subject(s) - computer science , association rule learning , scheme (mathematics) , overhead (engineering) , data mining , distributed database , information privacy , data sharing , secret sharing , distributed computing , cryptography , algorithm , computer security , mathematics , medicine , mathematical analysis , alternative medicine , pathology , operating system
In distributed computing, data sharing is inevitable, however, moving local databases from one site to another should be avoided because of the computational overhead and privacy consideration. Most of the data mining algorithms are designed assuming that data repository is stored locally. This paper presents a scheme and algorithms for mining association rules in geographically distributed data. The proposed scheme preserves data privacy of the different geographical site by passing secure messages between them. The algorithms minimize the communication cost by exchanging statistical summaries of the local databases. We provide a privacy and security analysis that shows the privacy preserving aspects of the proposed algorithms. Moreover, the paper presents extensive simulation experiments to evaluate the efficiency of the proposed scheme.