Error Types Of Machine Translated Academic Texts And Post-editing Strategies

The COVID-19 has had a significant impact on the global economy,politics,culture,society,life and other aspects.It not only endangers people’s physical health,but also causes damage to their mental health.The author selects the third chapter of the academic text,Mental Health Effects of COVID-19 to translate.This translation material mainly introduces the mental health problems and corresponding coping strategies of the front-line medical workers in Ghana and Uganda when dealing with the COVID-19 pandemic.The language is objective,rigorous,information intensive,coherent and logical.In recent years,neural network-based machine translation has advanced quickly.Machines “learn” from a big corpus of texts,which increases translation speed and lowers cost.Machine translation still has many problems,though.The major drawback is that the quality of translation is drastically reduced because it can only translate using the original form or literal meaning.Numerous academics are interested in finding ways to enhance the accuracy of machine translation,accelerate the translation process,and increase the productivity of translators.One of the best methods to increase translation efficiency without lowering the quality of the translation is by manual editing after machine translation.At present,many studies on machine translation and post-editing strategies have been carried out,but there is still a large research space for machine translation of specific text types,especially academic texts.Guided by Li Changshuan’s CEA analysis framework,the author lists common error types and corresponding feasible post editing strategies in machine translation of academic texts at the lexical,syntactic,and textual levels from the perspective of contrastive research between English and Chinese,and validates the effectiveness of the series of strategies through examples.