Identifying MT Errors for Higher-Quality Target Language Writing

Abstract

Second language education has arrived at a phase of proposing effective uses of neural machine translation (NMT). Previous research has explored various aspects of post-editing and suggested that it is crucial for producing better target-language (TL) texts. The purpose of this study is to identify MT errors in output text, so that L2 learners can recognize what to be aware of. The study targets MT translations of Japanese-written academic reports pre-edited by 73 Japanese students with intermediate TL (English) proficiency. The data were analysed to detect syntactical and semantical issues and systematically classify them. Results showed that the use of inappropriate TL vocabulary was the most frequent error, followed by redundant or repeated TL expressions. Some could be avoided in a pre-editing phase by carefully choosing precise source-language (SL) vocabulary, reducing SL ambiguity, while others required a deeper understanding of TL syntactic rules or the nuance of TL vocabulary. Learners need to raise their awareness of these MT issues to effectively conduct post-editing.