The evaluation of the quality of machine translation is a relevant and necessary study, because with the proliferation of subject matter related to innovation and technology, more and more people around the world are using neural machine translation. Therefore, in order to achieve a high quality of the translated text, the translation of the names of scientific concepts is particularly important. The purpose of the study is to assess the quality of machine translated artificial intelligence terms by performing a descriptive and comparative qualitative analysis of translation units. During the research, the quality of the translation of terms and pseudo-terms translated from English to Lithuanian with the neural machine translator “Google Translate” will be measured manually (Haque, R., Hasanuzzaman, M., Way, A. error typology), the most frequent inaccuracies and errors will be determined, a bilingual list of artificial intelligence terms and pseudo-terms based on examples of real usage and suggestions for improving the quality of terminology translation are made.
Project funding:
Project is funded by EU Structural Funds according to the 2014–2020 Operational Programme for the European Union Funds’ Investments priority “Development of scientific competence of researchers, other researchers, students through practical scientific activities” under Measure No. 09.3.3-LMT-K-712.
Project results:
The machine translation system Google Translate correctly translated the majority of AI terms and pseudo-terms, and the most frequently applied correct translation category was morphological variation. Most often, mistakes were made by not translating the term, and rarely by making a mistake in the order of words. In conclusion, we can say that the machine translation system Google Translate translates averagely, since the correct translation cases and errors are distributed almost evenly, so the overall translation quality when evaluating the translation of terms is evaluated positively.
Period of project implementation: 2021-09-01 - 2022-03-31
Project coordinator: Kaunas University of Technology
