Google Translate: Strategies to Optimize Its Academic Writing Translation

Abstract

Google Translate (GT) has been increasingly used by academics in translating their articles into English. Various translation errors are identified in the results produced by GT, limiting its functionality. Our present research had the objectives to: (1) identify errors produced in Indonesian-English GT translation and, based on the previous results, (2) explore strategies for optimizing Indonesian-English GT translation. This qualitative research, employing the Grounded Theory model, analyzed nine abstract documents (ca. 2100 words). There were 144 errors identified in the Indonesian-English GT translation of our documents, consisting of lexical errors (45%), grammatical errors (43.1%), and excessive wording (11.8%). Twelve strategies for modifying the Indonesian documents were proposed in order to assist GT producing better English documents. The strategies were grouped into Avoid, Maintain, and Enhance categories based on the nature of modification done to the documents. Detailed results are presented, followed by the discussion of how the proposed strategies might affect GT functionality and pedagogical practice.