Show simple item record

dc.contributor.advisorAranberri Monasterio, Nora
dc.contributor.advisorBorg, Claudia
dc.contributor.authorSarajlic, Jelena
dc.date.accessioned2023-06-30T14:34:36Z
dc.date.available2023-06-30T14:34:36Z
dc.date.issued2023-06-30
dc.identifier.urihttp://hdl.handle.net/10810/61811
dc.description.abstract[EN] This Master Thesis analyses the effect of neural machine translation on the language of the translation in terms of lexical, morphological, and syntactical diversity or richness. Four neural machine translation models are trained. Two different corpora of similar length and domain, one of which was created in this work, are used to train and evaluate the models, as well as translate text. Two language pairs were used in both directions: English and Spanish; and English and Croatian. Regarding lexical richness, the majority of our results indicate a degree of lexical loss in the translations. One metric shows a gain of lexical diversity in one of the translations. In morphological richness, the results are not as clear, with most of the metrics showing slight to no loss, or even a gain of richness in two of the translations. Part of speech distribution analysis, as well as parse distribution analyses, both seem to confirm claims made by some that neural machine translation systems increase the frequency of most and decrease the frequency of least frequent items.es_ES
dc.description.abstract[EU] Master-lan honen helburua da aztertzea itzultzaile automatiko neuronalek duten eragina proposatzen dituzten itzulpenetako hizkuntzan, aniztasun eta aberastasun lexikoari, morfologikoari eta sintaktikoari dagokionez. Horretarako, lau itzultzaile neuronal entrenatu dira. Entrenamendua, ebaluazioa eta itzulpen automatikoak egin dira domeinu eta luzera antzeko bi corpus erabilita (bata bereziki lan honetarako sortua). Bi hizkuntza-pare landu dira, noranzko bietan: ingelesa eta gaztelania batetik, ingelesa eta kroaziera bestetik. Aberastasun lexikoari erreparatuta, emaitza gehienek adierazten dute galera maila bat edo beste. Hala ere, metriketako batek aniztasun lexikoaren gehitzea gertatu izana erakusten du. Aberastasun morfologikoari buruz emaitzak ez dira argiak, izan ere, metrika gehienek galera txikia edo galerarik eza adierazten dute, eta, bi kasutan, aberastasunaren igoera. Kategoria gramatikalen eta sintaxiaren distribuzio-analisiari begiratuta, gure emaitzak bat datoz ikerlariek aurretiaz egindako baieztapenekin, hau da, itzultzaile neuronalek maiztasun handiko elementuen agerpenak areagotzen dituzte eta maiztasun gutxikoenenak mugatu.es_ES
dc.language.isoenges_ES
dc.rightsinfo:eu-repo/semantics/openAccess
dc.subjectmachine translationes_ES
dc.subjectlexical richnesses_ES
dc.subjectmorphological richnesses_ES
dc.subjectanalyzing translationses_ES
dc.subjectsyntactical diversityes_ES
dc.subjectlexical diversityes_ES
dc.subjectmorphological diversityes_ES
dc.subjecteffect of machine translationes_ES
dc.titleHow does machine translation affect language? Analyzing the effect of machine translation on translated textses_ES
dc.typeinfo:eu-repo/semantics/masterThesis
dc.date.updated2022-09-13T08:01:16Z
dc.language.rfc3066es
dc.rights.holder© 2022, la autora
dc.contributor.degreeMáster Universitario Erasmus Mundus en Tecnologías del Lenguaje y la Comunicación (LCT)
dc.contributor.degreeHizkuntzaren eta Komunikazioaren Teknologiak Erasmus Mundus Unibertsitate Masterra (LCT)
dc.contributor.degreeErasmus Mundus Master in Language and Communication Technologies (LCT)
dc.identifier.gaurregister126996-1096630-09es_ES
dc.identifier.gaurassign140447-1096630es_ES


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record