DC FieldValueLanguage
dc.contributor.authorDžunić, Zoranen
dc.contributor.authorMomčilović, Svetislaven
dc.contributor.authorTodorović, Branimiren
dc.contributor.authorStanković, Miomiren
dc.date.accessioned2020-12-11T13:04:36Z-
dc.date.available2020-12-11T13:04:36Z-
dc.date.issued2006-12-01en
dc.identifier.isbn1-4244-0433-9en
dc.identifier.urihttp://researchrepository.mi.sanu.ac.rs/handle/123456789/4382-
dc.description.abstractCorefence resolution is the process of determining whether two expressions in natural language refer to the same entity in the world. We adopt machine learning approach using decision tree to a coreference resolution of general noun phrases in unrestricted text based on well defined features. We also use approximate matching algorithms for a string match feature and databases of American last names and male and female first names for gender agreement and alias feature. For the evaluation we use MUC-6 coreference corpora. We show that pessimisitc error pruning method gives better generalization in a coreference resolution task than that reported in Soon et al. [20], when weights of positive and negative examples are properly chosen.en
dc.publisherIEEE-
dc.relation.ispartof8th Seminar on Neural Network Applications in Electrical Engineering, Neurel-2006 Proceedingsen
dc.subjectApproximate string matching | Coreference resolution | Decision tree | Machine learning | Pessimistic error pruningen
dc.titleConference resolution using decision treesen
dc.typeConference Paperen
dc.identifier.doi10.1109/NEUREL.2006.341188en
dc.identifier.scopus2-s2.0-46749112752en
dc.relation.firstpage109en
dc.relation.lastpage114en
item.openairetypeConference Paper-
item.cerifentitytypePublications-
item.fulltextNo Fulltext-
item.openairecristypehttp://purl.org/coar/resource_type/c_18cf-
item.grantfulltextnone-
Show simple item record

SCOPUSTM   
Citations

7
checked on Jun 11, 2024

Page view(s)

40
checked on May 9, 2024

Google ScholarTM

Check

Altmetric

Altmetric


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