Methods for Evaluating the Translation Quality of Artificial Intelligence Translator DEEPL Based on Multi Translation Parallel Corpus

With the acceleration of globalization and the development of information technology, the demand for cross linguistic communication is increasing day by day. The existing translation quality assessment methods have limitations in terms of accuracy and efficiency in evaluating large-scale data multilingual pairs. This article evaluated the translation quality of the artificial intelligence translator DeepL based on a multi translation parallel corpus. A multi translation parallel corpus based evaluation method was studied, which evaluated translation quality by comparing the similarity between DeepL translation results and the original text of multiple translation versions. The experimental results show that the method studied in this article has the highest similarity in Chinese translation for English, Russian, and French, reaching 98.7%, 98.9%, and 97.8%, respectively. The conclusion indicates that the DEEPL translation quality evaluation method based on multi translation parallel corpora can achieve accurate and efficient evaluation, which helps to improve multi-dimensional evaluation of translation quality.