Innovative Research on the Teaching Mode of English Translation Course in Colleges and Universities with the Support of Deep Learning

Abstract

Nowadays, teachers have begun to try to carry out teaching innovation, but there is still the phenomenon of copying the traditional course content and teaching form on the online teaching platform. Aiming at these types of teaching problems, this paper proposes research on the innovation of English translation teaching modes based on deep learning. In order to better analyze the effect of positional distance on word vectors, the PW-CBOW model is constructed on the basis of the structure of the CBOW model, and its parameters are optimized using the Adam optimizer. When the input and output are both indeterminate long sequences, a decoder-encoder is used for processing. The machine translation model is jointly constructed by combining the attention mechanism and the bidirectional gated loop unit. The English teaching model that combines the machine translation model is exemplified by using data analysis software. The results present that this paper’s model has +3.15 and +2.12 more BLEU points than the Transformer model and DynamicConv, respectively, and there is a significant difference between this paper’s teaching model and the traditional teaching model in the three dimensions of final grades, stage grades, and project grades, all of which satisfy P<0.05. A comprehensive analysis of the results suggests that this paper’s teaching model has a significant impact on students’ performance. The comprehensive analysis results show that the teaching model of this paper is of great significance to the improvement of student performance and the development of teaching in colleges.

Keywords

• pw-cbow model
• attention mechanism
• two-way gating cycle
• machine translation model
• english teaching mode
• 97d80