Show simple item record

dc.contributor.authorLøkse, Sigurd
dc.contributor.authorBianchi, Filippo Maria
dc.contributor.authorSalberg, Arnt-Børre
dc.contributor.authorJenssen, Robert
dc.date.accessioned2018-09-06T10:38:56Z
dc.date.available2018-09-06T10:38:56Z
dc.date.issued2017-05-19
dc.description.abstractIn this paper, we propose <i>PCKID</i>, a novel, robust, kernel function for spectral clustering, specifically designed to handle incomplete data. By combining posterior distributions of Gaussian Mixture Models for incomplete data on different scales, we are able to learn a kernel for incomplete data that does not depend on any critical hyperparameters, unlike the commonly used RBF kernel. To evaluate our method, we perform experiments on two real datasets. <i>PCKID</i>outperforms the baseline methods for all fractions of missing values and in some cases outperforms the baseline methods with up to 25% points.en_US
dc.descriptionManuscript version. The final publication is available at Springer via <a href=https://doi.org/10.1007/978-3-319-59126-1_36> https://doi.org/10.1007/978-3-319-59126-1_36</a>.en_US
dc.identifier.citationLøkse, S., Bianchi, F.M., Salberg, A.-B. & Jenssen, R. (2017). Spectral clustering using PCKID – A probabilistic cluster kernel for incomplete data. Lecture Notes in Computer Science, 10269 LNCS, 431-442. https://doi.org/10.1007/978-3-319-59126-1_36en_US
dc.identifier.cristinIDFRIDAID 1489865
dc.identifier.doi10.1007/978-3-319-59126-1_36
dc.identifier.issn0302-9743
dc.identifier.issn1611-3349
dc.identifier.urihttps://hdl.handle.net/10037/13697
dc.language.isoengen_US
dc.publisherSpringer Verlag (Germany)en_US
dc.relation.journalLecture Notes in Computer Science
dc.relation.projectIDinfo:eu-repo/grantAgreement/RCN/IKTPLUSS/239844/Norway/Next Generation Kernel-Based Machine Learning for Big Missing Data Applied to Earth Observation//en_US
dc.rights.accessRightsopenAccessen_US
dc.subjectVDP::Teknologi: 500::Informasjons- og kommunikasjonsteknologi: 550en_US
dc.subjectVDP::Mathematics and natural science: 400::Information and communication science: 420en_US
dc.subjectMissing dataen_US
dc.subjectRobustnessen_US
dc.subjectKernel methodsen_US
dc.subjectSpectral clusteringen_US
dc.titleSpectral clustering using PCKID – A probabilistic cluster kernel for incomplete dataen_US
dc.typeJournal articleen_US
dc.typeTidsskriftartikkelen_US
dc.typeManuskriptno
dc.typePeer revieweden_US
dc.typePreprinten_US


File(s) in this item

Thumbnail

This item appears in the following collection(s)

Show simple item record