Clustering with finite data from semi-parametric mixture distributions

Wang, Yong; Witten, Ian H.

Clustering with finite data from semi-parametric mixture distributions

dc.contributor.author	Wang, Yong
dc.contributor.author	Witten, Ian H.
dc.date.accessioned	2008-10-17T03:51:19Z
dc.date.available	2008-10-17T03:51:19Z
dc.date.issued	1999-11
dc.description.abstract	Existing clustering methods for the semi-parametric mixture distribution perform well as the volume of data increases. However, they all suffer from a serious drawback in finite-data situations: small outlying groups of data points can be completely ignored in the clusters that are produced, no matter how far away they lie from the major clusters. This can result in unbounded loss if the loss function is sensitive to the distance between clusters. This paper proposes a new distance-based clustering method that overcomes the problem by avoiding global constraints. Experimental results illustrate its superiority to existing methods when small clusters are present in finite data sets; they also suggest that it is more accurate and stable than other methods even when there are no small clusters.	en_US
dc.format.mimetype	application/pdf
dc.identifier.citation	Wang, Y. & Witten, H. (1999). Clustering with finite data from semi-parametric mixture distributions. (Working paper 99/14). Hamilton, New Zealand: University of Waikato, Department of Computer Science.	en_US
dc.identifier.issn	1170-487X
dc.identifier.uri	https://hdl.handle.net/10289/1043
dc.language.iso	en
dc.publisher	Dept. of Computer Science, University of Waikato	en_NZ
dc.relation.ispartofseries	Computer Science Working Papers
dc.subject	computer science	en_US
dc.subject	Machine learning
dc.title	Clustering with finite data from semi-parametric mixture distributions	en_US
dc.type	Working Paper	en_US
pubs.elements-id	55034
pubs.place-of-publication	Hamilton	en_NZ
uow.relation.series	99/14