The Review of Discourse-Based Machine Translation Evaluation

This article reviews the discourse-based machine translation evaluation metrics. According to different methods of using, machine translation evaluation based on discourse is divided into two categories: evaluation based on discourse structure and evaluation based on discourse features. For these two different categories, we introduce their representative works, and compare the advantages and disadvantages of each method. Finally, we summarize the evaluation metrics based on discourse, points out the problems to be paid attention to when evaluating, and predicts the future development trends.