Digital Transformation Paths in English Language Teaching from an Eco-Linguistic Perspective

Abstract

This paper follows the guiding principles of eco-linguistics for English teaching, improves the encoder of the question generation model, uses the gated self-attention mechanism to refine the textual information, obtains the embedded representations of the articles, fuses them with the embedded representations of the answers, and constructs a new automatic generation model of the English teaching exercises as a way of assisting in the digital transformation of the English classroom. After the controlled test, it was found that the post-test control class scored 71.954 points, and the experimental class scored 84.309 points, which was 12.355 points higher than the control class, with Sig. The two-tailed test resulted in a coefficient of 0.035, which was less than 0.05, and there was a significant difference in the post-test scores. In the survey of interest in learning, the average number of respondents who chose “strongly agree” or “agree” was 89%, 8.5% said “uncertain”, and 2.5% “disagreed”. Overall, the interest of the experimental class students in English learning has improved significantly. They not only pay more attention in class but also have a higher interest in practicing English after class. The digital transformation of traditional English classroom teaching is addressed in this study through the provision of new ideas and methods.

Keywords

• ecolinguistics
• natural language processing
• graph attention networks
• english language teaching
• 97p20