Two-level k-means clustering algorithm for k-tau relationship establishment and linear-time classification

Chitta, Radha and Murty, M Narasimha (2010) Two-level k-means clustering algorithm for k-tau relationship establishment and linear-time classification. In: Pattern Recognition, 43 (3). pp. 796-804.

PDF 26.pdf - Published Version Restricted to Registered users only Download (416kB) | Request a copy

Abstract

Partitional clustering algorithms, which partition the dataset into a pre-defined number of clusters, can be broadly classified into two types: algorithms which explicitly take the number of clusters as input and algorithms that take the expected size of a cluster as input. In this paper, we propose a variant of the k-means algorithm and prove that it is more efficient than standard k-means algorithms. An important contribution of this paper is the establishment of a relation between the number of clusters and the size of the clusters in a dataset through the analysis of our algorithm. We also demonstrate that the integration of this algorithm as a pre-processing step in classification algorithms reduces their running-time complexity.

Item Type:	Journal Article
Publication:	Pattern Recognition
Publisher:	Elsevier science
Additional Information:	Copyright for this article belongs to Elsevier science.
Keywords:	Clustering; k-Means; Classification; Linear-time complexity; Support vector machines; k-Nearest neighbor classifier
Department/Centre:	Division of Electrical Sciences > Electrical Engineering Division of Electrical Sciences > Computer Science & Automation
Date Deposited:	19 Jan 2010 08:57
Last Modified:	19 Sep 2010 05:54
URI:	http://eprints.iisc.ac.in/id/eprint/25429

Actions (login required)

View Item