A tool for measuring linguistic variations in machine translation: a corpus based study

Abstract EN

This paper is primarily a translation analysis of the Arabic and English morpho-syntactic structures using Biber's model
(1988) and Stanford computer program.

It is a corpus based quantitative study that has used 66 features out of the 67 that has been
identified by Biber and the paper is in line with Biber's model and statistical procedure.

The corpus selected for this thesis is Alice Monro's
collection of short stories The Power of Love (1985) and its translation into Arabic, Masiret El Hob (2015) by Mohamed Tantawi.

All the
English and Arabic features are counted by the aid of a computer program, Stanford (2015).

Stanford is a program that is used for
annotating the chosen corpus and it is working on the morpho-syntactic levels of English and Arabic.

The 66 features are classified into
four factors for the English language and five factors for the Arabic.

Only twelve features are counted manually in the Arabic analysis by
the researcher herself.

Finally, the findings reflect some differences between the two languages.