000190030 001__ 190030
000190030 005__ 20181203023314.0
000190030 0247_ $$2doi$$a10.1016/j.patcog.2014.10.003
000190030 022__ $$a0031-3203
000190030 02470 $$2ISI$$a000348880300037
000190030 037__ $$aARTICLE
000190030 245__ $$aCluster Validity Measure and Merging System for Hierarchical Clustering considering Outliers
000190030 269__ $$a2015
000190030 260__ $$bElsevier$$c2015
000190030 336__ $$aJournal Articles
000190030 520__ $$aClustering algorithms have evolved to handle more and more complex structures. However, measures allowing to qualify the quality of such partitions are rare and only specic to certain algorithms. In this work, we propose a new cluster validity measure (CVM) handling solutions with arbitrary shapes and various levels of outlier rejection based on notions of cluster cores and outliers. Moreover, we propose an adequate cluster merging system (CMS) to group cluster cores sharing some of their outliers. These outliers may be a mixture of these nearby cores. The extension of the Support Vector Clustering and Gaussian Process Clustering to obtain true hierarchical solutions are presented and applied using the proposed CVM and CMS in synthetic and real experiments showing the benefit for hyperparameter selection.
000190030 6531_ $$aclustering quality
000190030 6531_ $$ahierarchical clustering
000190030 6531_ $$aGaussian processes
000190030 6531_ $$aSupport Vector clustering
000190030 6531_ $$aKernels
000190030 6531_ $$aLTS5
000190030 700__ $$0242940$$g166738$$aDe Morsier, Frank
000190030 700__ $$0245927$$g150680$$aTuia, Devis
000190030 700__ $$0245002$$g167538$$aBorgeaud, Maurice
000190030 700__ $$aGass, Volker$$g217963$$0246190
000190030 700__ $$aThiran, Jean-Philippe$$g115534$$0240323
000190030 773__ $$j48$$tPattern Recognition$$k4$$q1478-1489
000190030 909C0 $$xU10954$$0252394$$pLTS5
000190030 909C0 $$pLASIG$$xU10244$$0252045
000190030 909C0 $$xU11066$$0252355$$pCTS
000190030 909CO $$particle$$pSTI$$pENAC$$ooai:infoscience.tind.io:190030
000190030 917Z8 $$x166738
000190030 917Z8 $$x148230
000190030 937__ $$aEPFL-ARTICLE-190030
000190030 973__ $$rNON-REVIEWED$$sPUBLISHED$$aEPFL
000190030 980__ $$aARTICLE