Show simple item record

dc.contributor.authorEasterling, David R.en_US
dc.contributor.authorWatson, Layne T.en_US
dc.contributor.authorRamakrishnan, Narenen_US
dc.contributor.authorHossain, M. Shahriaren_US
dc.date.accessioned2014-01-17T22:54:02Z
dc.date.available2014-01-17T22:54:02Z
dc.date.issued2013-12-31
dc.identifier.urihttp://hdl.handle.net/10919/24904
dc.description.abstractMany algorithms for constrained clustering have been developed in the literature that aim to balance vector quantization requirements of cluster prototypes against the discrete satisfaction requirements of constraint (must-link or cannot-link) sets. A significant amount of research has been devoted to designing new algorithms for constrained clustering and understanding when constraints help clustering. However, no method exists to systematically characterize solution sets as constraints are gently introduced and how to assist practitioners in choosing a sweet spot between vector quantization and constraint satisfaction. A homotopy method is presented that can smoothly track solutions from unconstrained to constrained formulations of clustering. Beginning the homotopy zero curve tracking where the solution is (fairly) well-understood, the curve can then be tracked into regions where there is only a qualitative understanding of the solution set, finding multiple local solutions along the way. Experiments demonstrate how the new homotopy method helps identify better tradeoffs and reveals insight into the structure of solution sets not obtainable using pointwise exploration of parameters.en_US
dc.language.isoenen_US
dc.publisherDepartment of Computer Science, Virginia Polytechnic Institute & State Universityen_US
dc.relation.ispartofComputer Science Technical Reportsen_US
dc.subjectNumerical analysisen_US
dc.subjectMachine learningen_US
dc.subjectScientific computingen_US
dc.subjectArtificial intelligenceen_US
dc.titleProbability-one Homotopy Maps for Constrained Clustering Problemsen_US
dc.typeTechnical reporten_US
dc.identifier.trnumberTR-13-09en_US
dc.type.dcmitypeTexten_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record