Title: On an adjacency cluster merit approach
Authors: Zeev Volkovich; Gerhard-Wilhelm Weber; Renata Avros; Orly Yahalom
Addresses: ORT Braude College of Engineering, Karmiel 21982, Israel. ' Institute of Applied Mathematics, Middle East Technical University, Ankara 06531, Turkey University of Siegen, Siegen 57076, Germany; University of Aveiro, Aveiro 3810-193, Portugal and Universiti Teknologi Malaysia, UTM Skudai, Johor 81310, Malaysia. ' ORT Braude College of Engineering, Karmiel 21982, Israel. ' ORT Braude College of Engineering, Karmiel 21982, Israel
Abstract: This work addresses the cluster validation problem of determining the 'right' number of clusters. We consider a cluster stability property based on the k-nearest neighbour type coincidences model. Quality of a clustering is measured by the deviation from this model, where a small deviation indicates a good clustering. The true number of clusters corresponds to the empirical deviation distribution having the shortest right tail. Experiments carried out on synthetic and real data sets demonstrate the effectiveness of our method.
Keywords: clustering; cluster stability; two-sample test; data mining; nearest neighbours; cluster validation.
International Journal of Operational Research, 2012 Vol.13 No.3, pp.239 - 255
Published online: 11 Jan 2015 *
Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article