Distributed Privacy Preserving Data Mining: A framework for k-anonymity based on feature set partitioning approach of vertically fragmented databases

JALPA, PATEL and KEYUR, RANA (2014) Distributed Privacy Preserving Data Mining: A framework for k-anonymity based on feature set partitioning approach of vertically fragmented databases. In: International Conference on Advances In Engineering And Technology - ICAET 2014, 24 - 25 May, 2014, RIT, Roorkee, India.

[img]
Preview
Text
20140726_112154.pdf - Published Version

Download (421kB) | Preview
Official URL: https://www.seekdl.org/conferences/paper/details/3...

Abstract

Recently, many data mining algorithms for discovering and exploiting patterns in data are developed and the amount of data about individuals that is collected and stored continues to rapidly increase. However, databases containing information about individuals may be sensitive and data mining algorithms run on such data sets may violate individual privacy. Also most organizations collect and share information for their specific needs very frequently. In such cases it is important for each organization to make sure that the privacy of the individual is not violated or sensitive information is not revealed. In this paper we have proposed a novel method to provide privacy to the data when the data is vertically partitioned and distributed over sites. In this work we presented trusted third party framework along with an application that generates k-anonymous dataset from two vertically partitioned sources without disclosing data from one site to other. K- anonymity constraint is satisfied using feature set partitioning method, which uses a genetic algorithm to search for optimal feature set partition and conventional asymmetric cryptographic technique will be used in case of trusted third party model. So in order to preserve privacy of the data trusted third party has been used and such data is first anonymized at local party using feature set partitioning method and then global classification and anonymization done at the trusted third party. We have proposed algorithm and tested different data sets for vertically partitions.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Distributed data mining; Privacy preserving; kanonymity; Genetic algorithm
Depositing User: Mr. John Steve
Date Deposited: 27 May 2019 04:41
Last Modified: 27 May 2019 04:41
URI: http://publications.theired.org/id/eprint/2717

Actions (login required)

View Item View Item