A Post-editing Report Of Computer-Aided Translation On Pharmaceutical Experiments(Excerpt)

This thesis is a translation practice report based on the translation of certain chapters of the textbook Pharmaceutical Experiments by the author.In the translation process,the model of machine translation+post editing is used and through analysis of the types of machine translation errors in pharmaceutical texts and the methods of modification in post-editing,the author could predict the parts with higher frequency of machine translation errors.It is envisaged to establish a pre-judgment system for post-translation editing,so as to improve the efficiency of post-translation editing in this field.The author finds that machine translation tends to be equivalent in form,while the source text,as a textbook,must not only ensure the accuracy of the information,but also realize its function/role in education,which means to achieve functional equivalence of the translation.According to the principle of Nida’s realization of functional equivalence,the author divides the difficulties caused by machine translation to post-editing in three levels:lexicon,syntax and semantics.The study found that a)the frequency of deficiencies at the lexical level is low,which can be revised by adding explanations or replacing relevant contextual words;b)deficiencies at the syntactic level need to break down long and complex sentences into easy-to-understand units,or adjust the sentence order in post editing process;c)misunderstandings at the semantic level require in-depth explanations of pharmaceutical operational terms or adjustments to the operating steps,to achieve the accuracy of the information received by the target readers.Focusing machine translation sentence errors during the process of post-translation editing,this article tentatively divides these errors into:accuracy problems detected on bilingual texts and fluency problems detected on target language texts.The analysis in this paper finds that according to Nida’s "kernels"model-analyzing the source language to extract the "kernel"→transfer→restructure in the target language,the functional equivalence of the translation can be realized.For accuracy issues,all the "kernels" in the sentence should be found in the post-editing process to determine the center of the sentence;for fluency issues,the relationship between multiple "kernels" should be clarified during restructuring,and other content should be made into attributive clauses,or the independent nominative structure enhances the readability of the translation.