Personalized Recommendations of Online English Teaching Resources for Higher Vocational Education

Abstract

Abstract Facing the intricate challenges of English instruction in higher vocational colleges, this research innovatively employs online resources to elevate educational quality and learner efficiency. It critiques the ineffectiveness of “one-size-fits-all” educational models and proposes a cutting-edge personalized recommendation system. This system adeptly curates learning materials to align with individual students’ learning patterns and competencies. By pioneering the TDINA and CUPMF models, and synergizing them with convolutional neural network (CNN) technology, our study dives deep into the educational content, ensuring precise resource recommendations. The system’s prowess was demonstrated through the successful recommendation of 815 resources, identifying strengths in electronic wall charts and courseware, alongside areas for enhancement in microclass scripts and recordings. A specialized experiment in personalized exercise recommendations further underscored the system’s efficacy, accurately tailoring the exercise level and difficulty to students’ learning capacities. Significantly, the CUPMF model emerged as a robust tool in refining the accuracy of these personalized recommendations, heralding a new era of bespoke educational experiences for students.