000865858 001__ 865858
000865858 005__ 20210130003202.0
000865858 0247_ $$2doi$$a10.1109/IJCNN.2019.8852146
000865858 0247_ $$2altmetric$$aaltmetric:69028841
000865858 037__ $$aFZJ-2019-05146
000865858 1001_ $$0P:(DE-Juel1)173704$$aMuzzarelli, Laura$$b0$$eCorresponding author$$ufzj
000865858 1112_ $$a2019 International Joint Conference on Neural Networks (IJCNN)$$cBudapest$$d2019-07-14 - 2019-07-19$$wHungary
000865858 245__ $$aRank Selection in Non-negative Matrix Factorization: systematic comparison and a new MAD metric
000865858 260__ $$bIEEE$$c2019
000865858 300__ $$a8
000865858 3367_ $$2ORCID$$aCONFERENCE_PAPER
000865858 3367_ $$033$$2EndNote$$aConference Paper
000865858 3367_ $$2BibTeX$$aINPROCEEDINGS
000865858 3367_ $$2DRIVER$$aconferenceObject
000865858 3367_ $$2DataCite$$aOutput Types/Conference Paper
000865858 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1573822801_22185
000865858 500__ $$aThis study was partly supported by the Helmholtz Portfolio Theme"Supercomputing and Modeling for the Human Brain" and the EuropeanUnion’s Horizon 2020 Research and Innovation Programme under GrantAgreement No. 785907 (HBP SGA2).
000865858 520__ $$aNon-Negative Matrix Factorization (NMF) is apowerful dimensionality reduction and factorization method thatprovides a part-based representation of the data. In the absence ofa priori knowledge about the latent dimensionality of the data, itis necessary to select a rank of the reduced representation. Severalrank selection methods have been proposed, but no consensusexists on when a method is suitable to use. In this work, we proposea new metric for rank selection based on imputation crossvalidation,and we systematically compare it against six othermetrics while assessing the effects of data properties. Usingsynthetic datasets with different properties, our work criticallyevidences that most methods fail to identify the true rank. Weshow that properties of the data heavily impact the ability ofdifferent methods. Imputation-based metrics, including our newMADimput, provided the best accuracy irrespective of the datatype, but no solution worked perfectly in all circumstances. Oneshould therefore carefully assess characteristics of their dataset inorder to identify the most suitable metric for rank selection.
000865858 536__ $$0G:(DE-HGF)POF3-574$$a574 - Theory, modelling and simulation (POF3-574)$$cPOF3-574$$fPOF III$$x0
000865858 536__ $$0G:(EU-Grant)785907$$aHBP SGA2 - Human Brain Project Specific Grant Agreement 2 (785907)$$c785907$$fH2020-SGA-FETFLAG-HBP-2017$$x1
000865858 588__ $$aDataset connected to CrossRef Conference
000865858 7001_ $$0P:(DE-Juel1)172811$$aWeis, Susanne$$b1$$ufzj
000865858 7001_ $$0P:(DE-Juel1)131678$$aEickhoff, Simon B.$$b2$$ufzj
000865858 7001_ $$0P:(DE-Juel1)172843$$aPatil, Kaustubh R.$$b3$$ufzj
000865858 773__ $$a10.1109/IJCNN.2019.8852146
000865858 8564_ $$uhttps://juser.fz-juelich.de/record/865858/files/Muzzarelli19.pdf$$yRestricted
000865858 8564_ $$uhttps://juser.fz-juelich.de/record/865858/files/Muzzarelli19.pdf?subformat=pdfa$$xpdfa$$yRestricted
000865858 909CO $$ooai:juser.fz-juelich.de:865858$$pec_fundedresources$$pVDB$$popenaire
000865858 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)173704$$aForschungszentrum Jülich$$b0$$kFZJ
000865858 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)172811$$aForschungszentrum Jülich$$b1$$kFZJ
000865858 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)131678$$aForschungszentrum Jülich$$b2$$kFZJ
000865858 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)172843$$aForschungszentrum Jülich$$b3$$kFZJ
000865858 9131_ $$0G:(DE-HGF)POF3-574$$1G:(DE-HGF)POF3-570$$2G:(DE-HGF)POF3-500$$3G:(DE-HGF)POF3$$4G:(DE-HGF)POF$$aDE-HGF$$bKey Technologies$$lDecoding the Human Brain$$vTheory, modelling and simulation$$x0
000865858 9141_ $$y2019
000865858 920__ $$lyes
000865858 9201_ $$0I:(DE-Juel1)INM-7-20090406$$kINM-7$$lGehirn & Verhalten$$x0
000865858 980__ $$acontrib
000865858 980__ $$aVDB
000865858 980__ $$aI:(DE-Juel1)INM-7-20090406
000865858 980__ $$aUNRESTRICTED