PRIVACY PRESERVING DATA MINING USING THRESHOLD BASED FUZZY C-MEANS CLUSTERING

Abstract
Privacy preserving is critical in the field of where data mining is transformed into cooperative task among individuals. In data mining, clustering algorithms are most skilled and frequently used frameworks. In this paper, we propose a privacy-preserving threshold clustering that uses code based technique with threshold estimation for sharing of secret data in privacy-preserving mechanism. The process includes code based methodology which enables the information to be partitioned into numerous shares and handled independently at various servers. The proposed method takes less number of iterations in comparison with existing methods that does not require any trust among the clients or servers. The paper additionally provides experimental results on security and computational efficiency of proposed method.

Authors
V Manikandan, V Porkodi, Amin Salih Mohammed, M Sivaram
Lebanese French University, Erbil, Iraqi Kurdistan

Keywords
Privacy Preserving, Data Mining, Threshold Cryptography, Fuzzy C-means Clustering, Vandermonde Matrix, Secure Multiparty Computation
Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 9 , Issue: 1 )
Date of Publication :
October 2018

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.