Show simple record

dc.contributor.authorMohaghegh, Mahsa
dc.contributor.authorSarrafzadeh, Hossein
dc.contributor.authorMohammadi, Mehdi
dc.date.accessioned2015-08-11T23:55:09Z
dc.date.available2015-08-11T23:55:09Z
dc.date.issued2014
dc.identifier.urihttps://hdl.handle.net/10652/2969
dc.description.abstractStatistical word alignment models need large amounts of training data while they are weak in small-sized corpora. This paper proposes a new approach of an unsupervised hybrid word alignment technique using an ensemble learning method. This algorithm uses three base alignment models in several rounds to generate alignments. The ensemble algorithm uses a weighed scheme for resampling training data and a voting score to consider aggregated alignments. The underlying alignment algorithms used in this study include IBM Model 1, 2 and a heuristic method based on Dice measurement. Our experimental results show that by this approach, the alignment error rate could be improved by at least 15% for the base alignment models.en_NZ
dc.language.isoenen_NZ
dc.publisherIEEE (Institute of Electrical and Electronics Engineers)en_NZ
dc.relation.urihttp://www.icmla-conference.org/icmla14/en_NZ
dc.relation.urihttp://www.researchgate.net/publication/272353675_Ensemble_Statistical_and_Heuristic_Models_for_Unsupervised_Word_Alignmenten_NZ
dc.subjectstatistical machine translation (SMT)en_NZ
dc.subjectstatistical word alignmenten_NZ
dc.subjectensemble learningen_NZ
dc.subjectheuristic word alignmenten_NZ
dc.titleEnsemble Statistical and Heuristic Models for Unsupervised Word Alignmenten_NZ
dc.typeConference Contribution - Paper in Published Proceedingsen_NZ
dc.rights.holderIEEE (Institute of Electrical and Electronics Engineers)en_NZ
dc.identifier.doi10.1109/ICMLA.2014.15en_NZ
dc.subject.marsden200323 Translation and Interpretation Studiesen_NZ
dc.identifier.bibliographicCitationMohaghegh, M., Sarrafzadeh, A., and Mohammadi, M. (2014). Ensemble Statistical and Heuristic Models for Unsupervised Word Alignment. The 13th International Conference on Machine Learning and Applications (ICMLA'14)(Ed.), Detroit, Michigan, USAen_NZ
unitec.institutionUnitec Institute of Technologyen_NZ
unitec.institutionWestern Michigan University ( Kalamazoo, Michigan, USA)en_NZ
unitec.publication.spage61en_NZ
unitec.publication.lpage66en_NZ
unitec.publication.titleMachine Learning and Applications (ICMLA), 2014 13th International Conference on Machine Learning and Applicationsen_NZ
unitec.conference.titleThe 13th International Conference on Machine Learning and Applications (ICMLA'14)en_NZ
unitec.conference.orgInternational Conference on Machine Learning and Applicationsen_NZ
unitec.conference.locationDetroit, Michigan, U.S.A.en_NZ
unitec.conference.sdate2014-12-03
unitec.conference.edate2014-12-05
unitec.peerreviewedyesen_NZ
unitec.identifier.roms57335en_NZ
unitec.identifier.roms56946


Files in this item

Thumbnail

This item appears in

Show simple record