Show simple item record

dc.contributor.authorCordeiro De Amorim, Renato
dc.contributor.authorMakarenkov, Vladimir
dc.contributor.authorMirkin, Boris
dc.date.accessioned2017-06-12T10:52:48Z
dc.date.available2017-06-12T10:52:48Z
dc.date.issued2016-11-20
dc.identifier.citationCordeiro De Amorim , R , Makarenkov , V & Mirkin , B 2016 , ' A-Ward pβ : Effective hierarchical clustering using the Minkowski metric and a fast k-means initialisation ' , Information Sciences , vol. 370-371 , pp. 343-354 . https://doi.org/10.1016/j.ins.2016.07.076
dc.identifier.issn0020-0255
dc.identifier.otherPURE: 11211479
dc.identifier.otherPURE UUID: a84426f0-feac-419e-8daa-347af7d2ed7f
dc.identifier.otherScopus: 84982851370
dc.identifier.urihttp://hdl.handle.net/2299/18316
dc.descriptionThis document is the Accepted Manuscript version of the following article: Renato Cordeiro de Amorin, Vladimir Makrenkov, and Boris Mirkin, 'A-Wardpβ: Effective hierarchical clustering using the Minkowski metric and a fast k-means initialisation', Information Services, Vol. 370-371, November 2016, pp. 343-354. The version of record is available online at doi: https://doi.org/10.1016/j.ins.2016.07.076.
dc.description.abstractIn this paper we make two novel contributions to hierarchical clustering. First, we introduce an anomalous pattern initialisation method for hierarchical clustering algorithms, called A-Ward, capable of substantially reducing the time they take to converge. This method generates an initial partition with a sufficiently large number of clusters. This allows the cluster merging process to start from this partition rather than from a trivial partition composed solely of singletons. Our second contribution is an extension of the Ward and Wardp algorithms to the situation where the feature weight exponent can differ from the exponent of the Minkowski distance. This new method, called A-Wardpβ, is able to generate a much wider variety of clustering solutions. We also demonstrate that its parameters can be estimated reasonably well by using a cluster validity index. We perform numerous experiments using data sets with two types of noise, insertion of noise features and blurring within-cluster values of some features. These experiments allow us to conclude: (i) our anomalous pattern initialisation method does indeed reduce the time a hierarchical clustering algorithm takes to complete, without negatively impacting its cluster recovery ability; (ii) A-Wardpβ provides better cluster recovery than both Ward and Wardp.en
dc.format.extent12
dc.language.isoeng
dc.relation.ispartofInformation Sciences
dc.rightsEmbargoed
dc.subjectFeature weighting
dc.subjectHierarchical clustering
dc.subjectInitialisation algorithm
dc.subjectMinkowski metric
dc.subjectControl and Systems Engineering
dc.subjectTheoretical Computer Science
dc.subjectSoftware
dc.subjectComputer Science Applications
dc.subjectInformation Systems and Management
dc.subjectArtificial Intelligence
dc.titleA-Wardpβ : Effective hierarchical clustering using the Minkowski metric and a fast k-means initialisationen
dc.contributor.institutionSchool of Computer Science
dc.description.statusPeer reviewed
dc.date.embargoedUntil2017-08-01
dc.identifier.urlhttp://www.scopus.com/inward/record.url?scp=84982851370&partnerID=8YFLogxK
dc.relation.schoolSchool of Computer Science
dc.description.versiontypeFinal Accepted Version
dcterms.dateAccepted2016-11-20
rioxxterms.versionAM
rioxxterms.versionVoR
rioxxterms.versionofrecordhttps://doi.org/10.1016/j.ins.2016.07.076
rioxxterms.licenseref.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/
rioxxterms.licenseref.startdate2017-08-01
rioxxterms.typeJournal Article/Review
herts.preservation.rarelyaccessedtrue
herts.date.embargo2017-08-01
herts.rights.accesstypeEmbargoed


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record