Report On Post-editing Machine Translation Of Energy Technology Perspectives 2020

With the rapid development of the language service industry,the traditional human translation has been unable to meet the increasingly diversified and complex needs of the market.However,the emergence of machine translation(hereinafter refer to as “MT”)technology has become one of the most effective ways to solve the language barrier.With high efficiency and low cost,MT can provide clients or readers with rough translation versions in a short time,making them grasp the main idea in advance.But nothing is perfect.MT version is difficult to be put into direct use or meet the personalized demand of clients for its poor readability and fluency.Therefore,there is a need to post-edit the MT version for the sake of better quality of translation and high acceptability for clients.In this report,the translator makes a pretranslation for the text “Energy Technology Perspectives 2020(Chapter one and two)” released by the International Energy Agency in July 2020 with Baidu translation and Google translation.By comparing the two MT versions,the translator summarizes the common problems that occurred to MT in light of“comprehension,expression and adaptation(CEA)”,and discusses how to make up for the shortcomings of MT with appropriate post editing(hereinafter refer to as “PE”)strategies based on a number of cases.At the comprehension level,the translator manages to solve the mistranslation via professional dictionaries,network resources,parallel text,which improves the accuracy of translation;at the expression level,substitution,conversion of part of speech,adjustment of word order are adopted to solve the inappropriate collocation,misuse of part of speech and semantic barrier in MT,which leads the translation to be more natural and authentic;at the adaptation level,amplification,simplification and the change on form or content to settle the under-translation or semantic redundancy encountered in MT,which enhances the readability and coherence of target text.It is hoped that this translation report can provide some reference for those who intend to deal with translation tasks in the way of human-computer interaction.