Investigating the Effect of Pre-editing on Machine Translation’s Final Output

Abstract

In a society with a state of the art technology and a globalised world with hundreds of languages, machine translation, as an interdisciplinary field, has become an integral part of foreign language learning. Its output quality, however, depends greatly on how one uses it. Hence, this research study primarily aims at investigating the effect of pre-editing on machine translation’s final product. Accordingly, it is hypothesised that pre-editing has a positive effect on machine translation’s final output; more precisely, there is a relationship between pre-editing quality and machine translation quality. To verify the aforementioned hypotheses, the researchers conducted an experimental based study with a set of 21 third year EFL students at the University of Mohamed Seddik Ben Yahia. This study was completed relying on a mixed-approach design, for the data were collected and analysed both qualitatively and quantitatively. The results obtained were very interesting. At first, the findings showed that most of the students were successful in applying the pre-editing rules and provided well-simplified texts. Second, the results corroborated the positive effect of pre-editing approach on machine translation’s intelligibility and fidelity. Lastly, the findings revealed that there is a significant relationship between pre-editing quality and machine translation quality. To conclude, the current study provides some useful recommendations and points out the necessity of teaching students the concept of pre-editing for better translation quality.