ViraScience.
Article

Entropy-based Consensus for Distributed Data Clustering

M. Owhadi-Kareshki, M.R. Akbarzadeh-T.

Abstract

The increasingly larger scale of available data and the more restrictive concerns on their privacy are some of the challenging aspects of data mining today. In this paper, Entropy-based Consensus on Cluster Centers (EC3) is introduced for clustering in distributed systems with a consideration for confidentiality of data; i.e. it is the negotiations among local cluster centers that are used in the consensus process, hence no private data are transferred. With the proposed use of entropy as an internal measure of consensus clustering validation at each machine, the cluster centers of the local machines with higher expected clustering validity have more influence in the final consensus centers. We also employ relative cost function of the local Fuzzy C-Means (FCM) and the number of data points in each machine as measures of relative machine validity as compared to other machines and its reliability, respectively. The utility of the proposed consensus strategy is examined on 18 datasets from the UCI repository in terms of clustering accuracy and speed up against the centralized version of FCM. Several experiments confirm that the proposed approach yields to higher speed up and accuracy while maintaining data security due to its protected and distributed processing approach.

Similar items

  • The "Best K" for Entropy-based Categorical Data Clustering

    Scholarly record2005SSDBMKeke Chen, Ling Liu

    With the growing demand on cluster analysis for categorical data, a handful of categorical clustering algorithms have been developed. Surprisingly, to our knowledge, none has satisfactorily addressed the important problem for categorical clustering – how can we determine the best

  • Distributed Data Clustering in Peer-to-Peer Networks: A Technical Review

    Scholarly record2014Rasool Azimi, Hedieh Sajedi

    Clustering as one of the main branches of data mining, has gained an important place in the different applied fields. On the other hand, Peer-to-Peer (P2P) networks with features such as simplicity, low cost communication, and high availability resources, have gained a worldwide

  • Distributed data clustering in multi-dimensional peer-to-peer networks

    Scholarly record2010ADCStefano Lodi, Gianluca Moro, Claudio Sartori

    Several algorithms have been recently developed for distributed data clustering, which are applied when data cannot be concentrated on a single machine, for instance because of privacy reasons or due to network bandwidth limitations, or because of the huge amount of distributed d

  • Consensus Big Data Clustering for Bayesian Mixture Models

    Open-access article2023Algorithms['Christos Karras', 'Aristeidis Karras', 'Konstantinos C. Giotopoulos', 'Markos Avlonitis', 'Spyros Sioutas', '']

    In the context of big-data analysis, clustering technique holds significant importance for effective categorization and organization extensive datasets. However, pinpointing ideal number clusters handling high-dimensional data can be challenging. To tackle these issues, several s

  • Distributed Data Clustering

    Scholarly record2003CAiSE Short Paper ProceedingsAbdelhamid Bouchachia

    To make effective use of distributed information, it is desirable to allow coordination and collaboration among various information sources. This paper deals with clustering data emanating from different sites. The process of clustering consists of three steps: find the (local) c

  • An Efficient Distributed Data Clustering Algorithm

    Scholarly record2010Malay K. Pakhira

    The k-means algorithm is one of the most popular clustering algorithms in use today. The high running time complexity of serial k-means limits its applicability for very large databases. On the other hand, the existing parallel kmeans algorithms demand huge data transfer operatio