Skip to main content
Article
Determining the Best K for Clustering Transactional Datasets: A Coverage Density-based Approach
Data and Knowledge Engineering
  • Hua Yan, Wright State University - Main Campus
  • Keke Chen, Wright State University - Main Campus
  • Ling Liu
Document Type
Article
Publication Date
1-1-2009
Abstract
The problem of determining the optimal number of clusters is important but mysterious in cluster analysis. In this paper, we propose a novel method to find a set of candidate optimal number Ks of clusters in transactional datasets. Concretely, we propose Transactional-cluster-modes Dissimilarity based on the concept of coverage density as an intuitive transactional inter-cluster dissimilarity measure. Based on the above measure, an agglomerative hierachical clustering algorithm is developed and the Merge Dissimilarity Indexes, which are generated in hierachical cluster merging processes, are used to find the candidate optimal number Ks of clusters of transactional data. Our experimental results on both synthetic and real data show that the new method often effectively estimates the number of clusters of transactional data.
Comments

The article posted is the authors' preprint version.

DOI
10.1016/j.datak.2008.08.005
Citation Information
Hua Yan, Keke Chen and Ling Liu. "Determining the Best K for Clustering Transactional Datasets: A Coverage Density-based Approach" Data and Knowledge Engineering Vol. 68 Iss. 1 (2009) p. 28 - 48 ISSN: 0169-023X
Available at: http://works.bepress.com/keke_chen/9/