Call For Paper Volume:4 Issue:8 Aug'2017 |

SECURED FREQUENT ITEMSET DISCOVERY IN MULTI PARTY DATA ENVIRONMENT

Publication Date : 31/12/2014



Author(s) :

M.P.ANITHA , A.SATHYAPRIYA , T.BALASUBRAMANIAM.


Volume/Issue :
Volume 1
,
Issue 6
(12 - 2014)



Abstract :

Security and privacy methods are used to protect the data values. Private data values are secured with confidentiality and integrity methods. Privacy model hides the individual identity over the public data values. Sensitive attributes are protected using anonymity methods. Two or more parties have their own private data under the distributed environment. The parties can collaborate to calculate any function on the union of their data. Secure Multiparty Computation (SMC) protocols are used in privacy preserving data mining in distributed environments. Association rule mining techniques are used to fetch frequent patterns.Apriori algorithm is used to mine association rules in databases. Homogeneous databases share the same schema but hold information on different entities. Horizontal partition refers the collection of homogeneous databases that are maintained in different parties. Fast Distributed Mining (FDM) algorithm is an unsecured distributed version of the Apriori algorithm. Kantarcioglu and Clifton protocol is used for secure mining of association rules in horizontally distributed databases. Unifying lists of locally Frequent Itemsets Kantarcioglu and Clifton (UniFI-KC) protocol is used for the rule mining process in partitioned database environment. UniFI-KC protocol is enhanced in two methods for security enhancement. Secure computation of threshold function algorithm is used to compute the union of private subsets in each of the interacting players. Set inclusion computation algorithm is used to test the inclusion of an element held by one player in a subset held by another.The system is improved to support secure rule mining under vertical partitioned database environment. The subgroup discovery process is adapted for partitioned database environment. The system can be improved to support generalized association rule mining process. The system is enhanced to control security leakages in the rule mining process.


No. of Downloads :

47


Indexing

Web Design MymensinghPremium WordPress ThemesWeb Development

SECURED FREQUENT ITEMSET DISCOVERY IN MULTI PARTY DATA ENVIRONMENT

December 11, 2014