<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>English translation teaching in colleges and universities has problems such as outdated models, lack of attractiveness, and low efficiency. In this paper, we construct an automatic translation scoring model for English teaching, convert the English translation scoring problem into a semantic similarity problem of multiple English phrases, combine the Bi-LSTM algorithm to realize lexical embedding encoding, and design a feature extraction algorithm mainly based on the Transformer encoder. The attention mechanism is introduced to interact with the English phrases with the translation similarity information and linguistic information, and the global optimal strategy is used to select the score as the final score of similarity and calculate the automatic translation scoring score. After the construction was completed, two classes of the same major in a university were used as the experimental class and the control class to conduct a controlled trial of the new teaching mode utilizing the model. It was found that after one semester of teaching, the translation scores of the experimental class and the control class were 94.63 and 82.77, respectively, and the gap between the two classes was 11.86 points, which was obvious compared to the pre-test gap. There is no significant change in the translation level of the control class, and the experimental class has made considerable progress, with its scores in the five dimensions being 12.7%, 6.7%, 26.6%, 13.9%, and 35.6% higher than those of the control class respectively. It can be concluded that the effectiveness of adopting the new teaching mode is remarkable. The students’ translation level has made tremendous progress, the model and the latest teaching mode are widely recognized and accepted by the students in the experimental class and produce greater attraction, and the students’ attitude towards translation learning is more positive. This study provides a useful exploration of the innovation of English translation teaching methods in colleges and universities and improves the efficiency and learning effect of English translation classrooms.</p>
</abstract>

English translation teaching in colleges and universities has problems such as outdated models, lack of attractiveness, and low efficiency. In this paper, we construct an automatic translation scoring model for English teaching, convert the English translation scoring problem into a semantic similarity problem of multiple English phrases, combine the Bi-LSTM algorithm to realize lexical embedding encoding, and design a feature extraction algorithm mainly based on the Transformer encoder. The attention mechanism is introduced to interact with the English phrases with the translation similarity information and linguistic information, and the global optimal strategy is used to select the score as the final score of similarity and calculate the automatic translation scoring score. After the construction was completed, two classes of the same major in a university were used as the experimental class and the control class to conduct a controlled trial of the new teaching mode utilizing the model. It was found that after one semester of teaching, the translation scores of the experimental class and the control class were 94.63 and 82.77, respectively, and the gap between the two classes was 11.86 points, which was obvious compared to the pre-test gap. There is no significant change in the translation level of the control class, and the experimental class has made considerable progress, with its scores in the five dimensions being 12.7%, 6.7%, 26.6%, 13.9%, and 35.6% higher than those of the control class respectively. It can be concluded that the effectiveness of adopting the new teaching mode is remarkable. The students’ translation level has made tremendous progress, the model and the latest teaching mode are widely recognized and accepted by the students in the experimental class and produce greater attraction, and the students’ attitude towards translation learning is more positive. This study provides a useful exploration of the innovation of English translation teaching methods in colleges and universities and improves the efficiency and learning effect of English translation classrooms.

An Innovative Model of English Translation Teaching Mode in Colleges and Universities from a Cross-cultural Perspective

Henan University of Animal Husbandry and Economy, School of Foreign Languages

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

English translation teaching in colleges and universities has problems such as outdated models, lack of attractiveness, and low efficiency. In this paper, we...

An Innovative Model of English Translation Teaching Mode in Colleges and Universities from a Cross-cultural Perspective

Online veröffentlicht: 05. Aug. 2024

Seitenbereich: -

Eingereicht: 31. März 2024

Akzeptiert: 20. Juni 2024

DOI: https://doi.org/10.2478/amns-2024-1977

Schlüsselwörter
Bi-LSTM, Multi-head attention, Automatic scoring, English teaching

© 2024 Sijia Zhang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

An Innovative Model of English Translation Teaching Mode in Colleges and Universities from a Cross-cultural Perspective

Online veröffentlicht: 05. Aug. 2024

Seitenbereich: -

Eingereicht: 31. März 2024

Akzeptiert: 20. Juni 2024

DOI: https://doi.org/10.2478/amns-2024-1977

SchlüsselwörterBi-LSTM, Multi-head attention, Automatic scoring, English teaching

© 2024 Sijia Zhang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Schlüsselwörter
Bi-LSTM, Multi-head attention, Automatic scoring, English teaching