Research on Translation Practice Based on Corpus and Data-Driven Learning

The role of translation quality evaluation in students' bilingual parallel corpus translation training is very important, but there is a problem of inaccurate outcome evaluation. Traditional translation teaching cannot solve the problem of translation quality evaluation in students' bilingual parallel corpus translation training, and the evaluation is unreasonable. Therefore, this paper proposes a fuzzy algorithm to optimize and innovate translation quality evaluation and analysis. Firstly, the interpretive theory is used to evaluate the translator, and the indicators are divided according to the requirements of translation quality evaluation to reduce the interference factors in translation quality evaluation. Then, the interpretation theory evaluates the translation quality of bilingual parallel corpus assisted translation teaching, forms a translation quality evaluation scheme, and comprehensively analyzes the translation quality evaluation results. MATLAB simulation shows that under certain evaluation criteria, the translation quality evaluation accuracy and translation efficiency of the fuzzy algorithm for bilingual parallel corpus-assisted translation teaching are better than those of traditional translation teaching.