Show simple item record

2016-11-23Diskussionspapier DOI: 10.18452/18450
Beta-boosted ensemble for big credit scoring data
dc.contributor.authorZieba, Maciej
dc.contributor.authorHärdle, Wolfgang Karl
dc.date.accessioned2017-10-09T13:36:36Z
dc.date.available2017-10-09T13:36:36Z
dc.date.issued2016-11-23
dc.identifier.urihttp://edoc.hu-berlin.de/18452/19129
dc.description.abstractIn this work we present a novel ensemble model for a credit scoring problem. The main idea of the approach is to incorporate separate beta binomial distributions for each of the classes to generate balanced datasets that are further used to construct base learners that constitute the final ensemble model. The sampling procedure is performed on two separate ranking lists, each for one class, where the ranking is based on prepotency of observing positive class. Two strategies are considered: one assumes mining easy examples and the second one forces good classification of hard cases. The proposed solutions are tested on two big datasets on credit scoring.eng
dc.language.isoeng
dc.publisherHumboldt-Universität zu Berlin
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/
dc.subjectcredit scoringeng
dc.subjectensemble modeleng
dc.subjectbeta distributioneng
dc.subjectBeta boosteng
dc.subjectbig dataeng
dc.subject.ddc330 Wirtschaft
dc.titleBeta-boosted ensemble for big credit scoring data
dc.typeworkingPaper
dc.identifier.urnurn:nbn:de:kobv:11-110-18452/19129-1
dc.identifier.doihttp://dx.doi.org/10.18452/18450
local.edoc.container-titleSonderforschungsbereich 649: Ökonomisches Risiko
local.edoc.pages21
local.edoc.type-nameDiskussionspapier
local.edoc.institutionWirtschaftswissenschaftliche Fakultät
local.edoc.container-typeseries
local.edoc.container-type-nameSchriftenreihe
local.edoc.container-volume2016
local.edoc.container-issue52
local.edoc.container-erstkatid2195055-6

Show simple item record