A Mobile English Learning Platform Based on Data Mining and Personalized Recommendations

As mobile technology advances, mobile English learning platforms have increasingly become the predominant mode of language acquisition. However, traditional platforms often suffer from inefficiency due to a lack of targeted and personalized recommendations. This paper explores the enhancement of mobile English learning platforms through the application of customized recommendation algorithms and data mining techniques. We have developed a system architecture composed of a user layer, a business layer, and a data layer, employing Spark and Support Vector Machine (SVM) for parallel data processing. Our approach integrates a collaborative filtering algorithm and a hybrid recommendation model, utilizing user information alongside the dual-tower model Deep Structured Semantic Model (DSSM) and the Deep Interest Network (DIN). Experimental results demonstrate that our recommendation model significantly surpasses traditional models in precision, recall, and F1 scores, with the F1 score improving by approximately 0.1 to 0.4. Furthermore, learners using this platform showed superior outcomes in homework completion, test scores, learning quality, and sustained learning enthusiasm compared to a control group, with an overall improvement of 15.48 points in English proficiency scores. In summary, this study validates the effectiveness of a mobile English learning platform that synergizes data mining with personalized recommendation algorithms to enhance both learning efficiency and user satisfaction.