—Due to explosion in the number of autonomous data sources, there is an emergent need for effective approaches to distributed clustering. Intuitionistic Fuzzy Set is a suitable tool to cope with imperfectly defined facts and data, as well as with imprecise knowledge. This paper introduces a novel intuitionistic fuzzy based distributed clustering algorithm, to cluster distributed datasets, without necessarily downloading all the data into a single site. The process is carried out in two different levels: local level and global level. In local level, numerical datasets are converted into intuitionistic fuzzy data and they are clustered independently from each other using modified fuzzy C-Means algorithm. In global level, global centroid is computed by clustering all local cluster centroids. The global centroid is again transmitted to local sites to update the local cluster model. The new algorithm is compared against two existing ensemble based distributed clustering algorithms and centralized clustering where all the data are merged into a single data source and clustered. The simulated experiments described in this paper confirm good performance of the proposed algorithm.
—Distributed Clustering, Fuzzy C-Means, Local Centroid, Global Centroid, Intuitionistic Fuzzy Sets.
N. Karthikeyani Visalakshi is with the Department of Computer Science, Vellalar College for Women, Erode – 638 012, Tamilnadu, India (phone: 91-424-2431936, e-mail: firstname.lastname@example.org).
K. Thangavel is with the Department of Computer Science, Periyar University, Salem – 636 011, Tamilnadu, India (e-mail: email@example.com).
R.Parvathi is with the Department of Mathematics, Vellalar College for Women, Erode – 638 012, Tamilnadu, India (e-mail: firstname.lastname@example.org).
Cite: N. Karthikeyani Visalakshi, K. Thangavel and R. Parvathi, "An Intuitionistic Fuzzy Approach to Distributed Fuzzy Clustering," International Journal of Computer Theory and Engineering
vol. 2, no. 2, pp. 295-302, 2010.