Englishafaan Oromoo Machine Translation An Experiment Using Statistical Approach

Machine Translation (MT) refers to the use of a machine for performing translationrntask which converts text in one Natural Language into another Natural Language. Itrncan have many applications like cross-linguistic information retrieval and speech tornspeech translation systems. It can also assist professional translators by producingrndraft quality output that reduces cost that would be incurred if translation and typingrnwas done manually from scratch.rnEnglish is the lingua franca of online information and Afaan Oromoo is one of thernmost resource scarce languages. For this reason, monolingual Afaan Oromoornspeakers need to use documents written in other languages, among which English isrnthe most popular one. To satisfy this need, translation of the English documents tornAfaan Oromoo, and thus, making these online documents available in Afaan Oromoornis vital in addressing the language barrier thereby reducing the effect of digital divide.rnTherefore, this thesis is focused on the development of a prototype English-AfaanrnOromoo machine translation system using statistical approach, i.e, without explicitrnformulation of linguistic rules, as this approach involves low cost and swiftest wayrnavailable these days. Using limited corpus of about 20, 000 bilingual sentences, arntranslation accuracy of 17.74% was achieved.

Get Full Work

Report copyright infringement or plagiarism