This paper introduces the state-of-the-art MT evaluation survey that contains both manual and automatic evaluation methods. The traditional human evaluation criteria mainly include the intelligibility, fidelity, fluency, adequacy, comprehension, and informativeness. We classify the automatic evaluation methods into two categories, including lexical similarity and linguistic features application. The lexical similarity methods contain edit distance, precision, recall, and word order, etc. The linguistic features can be divided into syntactic features and semantic features. Subsequently, we also introduce the evaluation methods for MT evaluation and the recent quality estimation tasks for MT. Machine Translation Evaluation: A Survey

Advertisements