Title: Hybrid privacy preserving clustering for big data while ensuring security

Authors: T.P. Pushphavathi; P.V.R. Murthy

Addresses: Ramaiah University of Applied Sciences (RUAS), # 470-P Peenya Industrial Area, 4th Phase, Bengaluru-560058, India ' Ramaiah University of Applied Sciences (RUAS), # 470-P Peenya Industrial Area, 4th Phase, Bengaluru-560058, India

Abstract: Big data is a recent technology in modern generation. Data are important information and analysed to ensure proper security. Data clustering is a fundamental technique in knowledge discovery and data engineering. Clustering a data using different algorithms is wildly used in various applications, such as soft computing, mobile computing and medicine. The traditional probabilistic clustering using c-means (PCM) algorithm is used in image analysis, and high dimensional data. However, it is difficult for PCM to produce a better result for clustering big data, mainly for heterogeneous data, since it is initially designed for small structured datasets. To contest this problem, it proposes a high-order PCM algorithm (HOPCM) for big data clustering by optimising the objective function in the tensor space. Further, we designed and developed a privacy preserved distributed HOPCM method based on MapReduce for very large amounts of heterogeneous data. The proposed approach of privacy preserving c-means (PPCM) combined with Brakerski-Gentry-Vaikuntanathan (BGV) shows better results for data security. Finally we compared the performance analysis of high-order PCM algorithm (HOPCM) with distributed high-order PCM algorithm (DHOPCM) and privacy preserving c-means (PPCM). The experimental results showed that HOPCM directs efficiently cluster a large number of heterogeneous data to provide security for private data.

Keywords: big data cluster; cloud computing; privacy preserving c-means; PPCM; high order PCM; HOPCM; distributed high-order possiblistic c-means; DHOPCM; Brakerski-Gentry-Vaikuntanathan; BGV.

DOI: 10.1504/IJCC.2021.119196

International Journal of Cloud Computing, 2021 Vol.10 No.4, pp.370 - 389

Received: 20 Jun 2019
Accepted: 28 Mar 2020

Published online: 25 Nov 2021 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article