Principles Of Task Development For Training Post-Editing Of Machine Translation

Abstract

Quality of machine translation increased rapidly and this led to the emergence of the profession of a post-editor. Post-editing of machine translation (PEMT) is an activity of improving machine translation and is in high demand in the translation market. However, there is no PEMT training program at Russian universities, so some specialists post-edit intuitively and apply machine translation to inappropriate types of texts. We aimed at developing PEMT training tasks based on the criteria and requirements for post-editors that we compiled as a result of self-observation and a survey of post-editing professionals. Our action research employs several stages of task development and such methods as comparative analysis and statistical analysis. The proposed tasks were tested on third-year students of Novosibirsk State Technical University who major in translation. While creating tasks, we used text on topics in demand with language service providers (LSP) clients. Based on the results, we offered a list of principles, which would be helpful for producing further tasks and teaching PEMT program at Russian universities.