Research on the Innovation of English Teaching Mode in Colleges and Universities under the Background of Digital Transformation

The digitalization of education promotes the transformation and development of education, and in the teaching of English in colleges and universities, information technology and digital technology provide teachers with brand-new teaching methods and platforms. Therefore, this paper constructs a set of multimodal knowledge methods for English courses based on knowledge mapping. The article firstly researches the fundamental theory and basic framework of knowledge graph, and proposes an entity recognition model based on RoBERTa-wwm-BiGRU-GlobalPointer, then carries out relationship extraction and multimodal entity linking, and finally builds a multimodal retrieval platform for courses based on knowledge graph. The article takes the English teaching classroom of a university as the research object to test and validate the teaching effect and teaching satisfaction of the English teaching mode based on knowledge graph, and the comparison of the difference between the pre-test means of the experimental class and the control class is 8.78 (post-test) > 6.99 (pre-test), and the comparison of the difference between the pre- and post-test means of the experimental class and the control class is 7.1 (experimental class) > 5.31 (control class). In the analysis of the covariance between the pre- and post-test academic performance of the experimental and control classes, the significant p-value of the pre-test performance is 0.000, which is less than 0.05, and the significant p-value of the teaching mode is 0.000, which is less than 0.05, which suggests that the English teaching mode based on the Knowledge Graph has a more significant impact on the students’ academic performance.