The Application Of Pre-editing Strategies In Machine Translation

In the last decade,Machine Translation(MT)has progressed significantly and has therefore become highly prevalent in the translation industry as a solution to meet increasing demand for tighter deadlines and to reduce translation costs.Nevertheless,the output produced by most engines is still less than satisfactory.In order to maximize the advantages of MT and human translation,MT+editing modes have gained more popularity.As one type of MT+editing mode,pre-editing mode has served as an important way to improve translation quality and efficiency.This thesis seeks to study the implication of pre-editing strategies in machine translation with the aim to improve both the efficiency and quality of translation.By taking the translation of Global Cities Forum as an example and adopting a comparative analysis,this thesis has made a text type analysis,and massive problems have been found in the source text including logic gap and redundancy.Apart from this,this thesis also discusses the problems of the text that would cause problems in the output of machine translation including logic misses,ambiguity and syntactical errors.With cues offered by Federal Plain Language Guide,this thesis has put forward applicable pre-editing strategies including addition or deletion,paraphrasing and explicitation in order to make the whole text more coherent and readable.Lastly,this thesis will apply these strategies to the pre-editing of the original text,and by making a comparison of the source text before pre-editing and after pre-editing,finally demonstrate that the application of pre-editing strategies makes the original text more readable and thus raises the efficiency and quality of machine translation.This thesis hopes to provide some useful hindsight on how to improve machine translation output.