Show simple item record

dc.contributor.authorCarvalho, Francisco de A.T. de
dc.contributor.authorBertrand, Patrice
dc.contributor.authorSimões, Eduardo C.
dc.date.accessioned2017-03-15T15:10:58Z
dc.date.available2017-03-15T15:10:58Z
dc.date.issued2016
dc.identifier.issn0925-2312
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/16363
dc.language.isoenen
dc.subjectSelf-organizing mapsen
dc.subjectBatch training algorithmsen
dc.subjectInterval-valued dataen
dc.subjectAdaptive distancesen
dc.subjectSymbolic data analysisen
dc.subject.ddc518en
dc.titleBatch SOM algorithms for interval-valued data with automatic weighting of the variablesen
dc.typeArticle accepté pour publication ou publié
dc.description.abstractenInterval-valued data is most utilized to represent either the uncertainty related to a single measurement, or the variability of the information inherent to a group rather than an individual. In this paper, we focus on Kohonen self-organizing maps (SOMs) for interval-valued data, and design a new Batch SOM algorithm that optimizes an explicit objective function. This algorithm can handle, respectively, suitable City-Block, Euclidean and Hausdorff distances with the purpose to compare interval-valued data during the training of the SOM. Moreover, most often conventional batch SOM algorithms consider that all variables are equally important in the training of the SOM. However, in real situations, some variables may be more or less important or even irrelevant for this task. Thanks to a parameterized definition of the above-mentioned distances, we propose also an adaptive version of the new algorithm that tackles this problem with an additional step where a relevance weight is automatically learned for each interval-valued variable. Several examples with synthetic and real interval-valued data sets illustrate the usefulness of the two novel batch SOM algorithms.en
dc.relation.isversionofjnlnameNeurocomputing
dc.relation.isversionofjnlvol182en
dc.relation.isversionofjnldate2016
dc.relation.isversionofjnlpages66-81en
dc.relation.isversionofdoi10.1016/j.neucom.2015.11.084en
dc.relation.isversionofjnlpublisherElsevieren
dc.subject.ddclabelModèles mathématiques. Algorithmesen
dc.relation.forthcomingnonen
dc.relation.forthcomingprintnonen
dc.description.ssrncandidatenonen
dc.description.halcandidatenonen
dc.description.readershiprechercheen
dc.description.audienceInternationalen
dc.relation.Isversionofjnlpeerreviewedouien
dc.relation.Isversionofjnlpeerreviewedouien
dc.date.updated2017-03-09T15:10:36Z
hal.person.labIds
hal.person.labIds
hal.person.labIds


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record