The Application of Social Media in College English Teaching: Data Analysis and Visualization Research

In the era of fragmented learning, using social media for English learning has become mainstream. College English teaching requires continuous understanding of students' online and offline learning situations and needs. Information such as the English learning topics discussed by students on social media and the areas of learning they focus on will provide important references for teachers in planning teaching and selecting teaching cases. This paper collects relevant college student English topic information from the Weibo platform and applies the GPU-DMM (Graphic Processing Unit-Dirichlet Multinomial Mixtures) semantically enhanced topic model to perform topic mining on English topics. It obtains the corresponding document-topic distribution and topic-word distribution, analyzes the differences between different topics, and identifies the focus points of college students in English learning. Finally, the information of different topics is visually displayed to systematically reflect the topical characteristics of college student English discussions.