Korreferentzia-ebazpena euskaraz idatzitako testuetan
No Thumbnail Available
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Artikulu honetan euskarazko korreferentziak ebazteko sistema baten garapena azaltzen da.Leheniketa behin, azterketa linguistiko batean oinarritutako aipamen-detektatzaile automatikoa aurkezten dugu.Sistema hori erregelatan oinarritutakoa da, eta egoera finituko teknologia erabiliz inplementatu da.Behin testuko aipamenak detektatuta, beraien artean gertatzen diren korreferentzia-erlazioak ebaztendituen sistema ere garatu dugu. Horretarako, ingeleserako diseinatu den sistema eraginkor bat oinarritzathartu, eta euskararen ezaugarrietara egokitu da. Egokitze-lan hori deskribatzen da artikuluaren bigarren partean.
This paper presents the first steps in the development of a Basque coreference resolution system. Firstly, wedescribe a mention detector system based on a linguistic study of the nature of mentions. The mention detectoris rule-based and has been implemented using finite state technology. The system identifies mentions that arepotential candidates to be part of coreference chains in Basque written texts. Finally, we describe the process ofadapting a state-of-the-art English coreference resolution system to Basque.
This paper presents the first steps in the development of a Basque coreference resolution system. Firstly, wedescribe a mention detector system based on a linguistic study of the nature of mentions. The mention detectoris rule-based and has been implemented using finite state technology. The system identifies mentions that arepotential candidates to be part of coreference chains in Basque written texts. Finally, we describe the process ofadapting a state-of-the-art English coreference resolution system to Basque.
Description
Keywords
aipamen-detekzioa, korreferentzia-ebazpena, hizkuntzaren prozesamendua, mention detection, coreference resolution, natural language processing