Analyzing Indonesian-English Abstracts Translation in view of Translation Errors by Google Translate (Published)
This study seeks to investigate the frequency of errors in the translation of abstracts produced by Google Translate with reference to Keshavarzʼs (1999) model of error analysis. This research will be of great benefit to undergraduate students to use these findings as a guideline in writing a thesis abstract. Five types of error classification is used as the parameters, namely lexicosemantic error, tense error, preposition error, word order error, distribution and use of verb group error, and active and passive voice error. The data were obtained from several faculties at the Methodist University of Indonesia, Medan. A total of ten abstracts of undergraduate students’ paper from various faculties were randomly selected. The data are then compared on each sentence segment and any words or phrases found to have errors are analyzed. The study revealed that 21 frequencies in terms of lexicosemantic errors, 9 frequencies in terms of tense errors, 13 frequencies in terms of preposition error, 27 frequencies in terms of word order error, 15 frequencies in terms of distribution and use of verb group errors, 8 frequencies in terms of active and passive voice errors.
Keywords: Abstract Translation, Error Analysis, Machine Translation, Translation Errors