Hybrid Approach to Speed-Up the Privacy Preserving Kernel K-means Clustering and its Application in Social Distributed Environment

Abstract

In this most revolutionized world, the social network plays a vital role in each and everyone’s life. Social networking is a pervasive communication platform where the users can search whole over the world via the Internet. Users have similar interest to connect and interact with one another and to share their private and personal interest. In this paper, we examine privacy concern for the social networking users by distributed clustering method. In the proposed scheme, to speed-up, the Kernel k-means algorithm, a prototype based hybrid kernel k-means algorithm is involved in distributing the users into the cluster. Since we are using a large data set, we use a hybrid approach to speed-up the kernel k-means clustering (HSKK). The clustering process used here is to partition a similar set of objects in a dataset. Additionally, in the clustering process, a cryptographic protocol such as homomorphic encryption is involved in every dataset to achieve the goal to protect the private data. To prove the efficiency of the proposed approach, the experiment is done on Movie lens dataset. The experimental study of HSKK shows that the proposed method can significantly reduce the computation time and the private data of users is hidden from the service provider.