Please use this identifier to cite or link to this item: https://open.ni.ac.rs/handle/123456789/4
DC FieldValueLanguage
dc.contributor.authorMančev, Dejanen
dc.contributor.authorTodorovic B.en
dc.date.accessioned2020-02-13T07:34:26Z-
dc.date.available2020-02-13T07:34:26Z-
dc.date.issued2012-12-01en
dc.identifier.isbn9781467315722en
dc.identifier.urihttps://open.ni.ac.rs/handle/123456789/4-
dc.description.abstractThe paper presents the use of a two structural model committee, where the output of the first model together with its confidence is set as the input of the second model. The confidence for the given context of predictions in the sequence is extracted from the alternative hypotheses generated from the first model. We present experiments on the shallow parsing, comparing the performance of the proposed method to the separate models. © 2012 IEEE.en
dc.relation.ispartof11th Symposium on Neural Network Applications in Electrical Engineering,NEUREL 2012 - Proceedingsen
dc.titleConfidence based learning of a two-model committee for sequence labelingen
dc.typeConference Paperen
dc.identifier.doi10.1109/NEUREL.2012.6419998en
dc.identifier.scopus2-s2.0-84874373187en
dc.identifier.urlhttps://api.elsevier.com/content/abstract/scopus_id/84874373187en
dc.contributor.orcid#NODATA#en
dc.contributor.orcid#NODATA#en
dc.relation.firstpage167en
dc.relation.lastpage170en
item.grantfulltextnone-
item.fulltextNo Fulltext-
crisitem.author.deptDepartman za računarske nauke-
crisitem.author.parentorgPrirodno-matematički fakultet-
Appears in Collections:Naučne i umetničke publikacije
Show simple item record

SCOPUSTM   
Citations

2
checked on Aug 3, 2020

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.