Exploring Google Translate-friendly strategies for optimizing the quality of Google Translate in academic writing contexts

The present study explored the types of errors found in Google Translate (GT) Chinese-to-English translations and, based on those error types, proposes strategies for optimizing the performance of GT. Seven abstracts written in both Chinese and English from seven articles published in English Teaching and Learning in 2017 were selected as the materials. The researchers compared the GT translations to the English abstracts written by the original author(s) and analyzed the problems in the translations. The problematic translations consisted of grammatical errors (35%) and lexical errors (65%). Relatedly, we propose nine specific strategies to employ when writing Chinese abstracts to be translated into English using GT. According to the strategies, we suggest that users (1) avoid native language-specific expressions, (2) maintain the use of original English terminologies in composing Chinese abstracts, and (3) enhance logical relations and expressions for the discipline-specific discourse community. Further analyses revealed that 99% of the 69 identified problems in the GT translations of the seven abstracts could be avoided by using the proposed strategies. A conceptual framework for the collaboration between GT and GT users is proposed and pedagogical implications are discussed.