Singapore University of Social Sciences

Machine Translation Post-Editing

Machine Translation Post-Editing (TNT505)

Synopsis

TNT505 Machine Translation Post-Editing will start with an overview of the development of machine translation before delving into the recent breakthroughs in the Neural Machine Translation. The course will briefly explain the algorithm behind the NMT and the impact of AI. Such knowledge will be useful for students to understand the strengths and weaknesses of different types of MT engines. Functions of MT pre-editing and post-editing will be discussed. Students will learn to identify where and how MT solutions are viable, considering specific business needs, budget and timeline. They will then focus on the skills of post-editing MT output based on a set of pre-established criteria to achieve a desired level of quality. At the end of the course, students will reflect on major challenges they face in MT post-editing and evaluate the efficiency in such an approach. The coursework will be project-based so that students will be assigned to practitioners of their language pairs as their project supervisors.

Level: 5
Credit Units: 5
Presentation Pattern: Every January
E-Learning: BLENDED - Learning is done MAINLY online using interactive study materials in Canvas. Students receive guidance and support from online instructors via discussion forums and emails. This is supplemented with SOME face-to-face sessions. If the course has an exam component, This will be administered on-campus.

Topics

  • The development of machine translation
  • State-of-the-art of Neural Machine Translation
  • Pre-editing vs post-editing processes
  • Light and full MT post-editing
  • Common MT errors
  • MT strategy and consulting

Learning Outcome

  • Compare different types of MT engines
  • Design workflow of human-aided machine translation
  • Prioritise different tasks in MT post-editing
  • Formulate an approach to the use of MT
  • Predict MT output after pre-editing
  • Appraise MT output
  • Improve the efficiency of the current design and workflow
Back to top
Back to top