Skip navigation
Please use this identifier to cite or link to this item: http://arks.princeton.edu/ark:/88435/dsp01bn999692c
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorFellbaum, Chrstiane-
dc.contributor.authorTampakis, Nikitas-
dc.date.accessioned2014-07-17T19:30:09Z-
dc.date.available2014-07-17T19:30:09Z-
dc.date.created2014-05-
dc.date.issued2014-07-17-
dc.identifier.urihttp://arks.princeton.edu/ark:/88435/dsp01bn999692c-
dc.description.abstractMachine translation (MT) is the use of computer software to translate bodies of texts across languages. State of the art MT systems facilitate global communication and aid in making human languages universally accessible. Google Translate is the de facto platform used for instant machine-based translations of texts. Although the statistical methods used by the system generally produce fairly reliable translation results, they are not designed to handle poetry. Since most automatic translation systems are guided by already existing parallel versions of texts written in multiple languages, they only work well with previously observed language. Poetry is a challenge to translate because it is formed differently from standard language. Observe the following excerpt from the Axion Esti, written by Greek Nobel Laureate poet, Odysseas Elytis. The original Greek is accompanied by a Google Translate result as well as a translation by scholar Edmund Keeley. The MT result neither makes sense nor is poetic.en_US
dc.format.extent42 pages*
dc.language.isoen_USen_US
dc.titleComputational Poetry: Translating Greek and Italian Popular Songsen_US
dc.typePrinceton University Senior Theses-
pu.date.classyear2014en_US
pu.departmentComputer Scienceen_US
pu.pdf.coverpageSeniorThesisCoverPage-
Appears in Collections:Computer Science, 1988-2020

Files in This Item:
File SizeFormat 
Tampakis_Nikitas_Thesis.pdf1.1 MBAdobe PDF    Request a copy


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