With the purpose of meeting the personalized needs of higher vocational students and optimizing the English vocabulary learning path, a personalized recommendation learning system for English vocabulary is designed based on big data. The improved K-means algorithm is used for clustering analysis of learner features, and the English vocabulary hybrid recommendation model is constructed by combining the user-based collaborative filtering recommendation algorithm and the item-based collaborative filtering recommendation algorithm. Then the questionnaire is used to evaluate the English vocabulary personalized recommendation learning system. The sample learners are categorized into six types of different English learning levels, with intermediate level students accounting for the largest proportion of 31.4%. The hybrid recommendation model in this paper has a certain improvement compared with the single collaborative filtering recommendation algorithm, with MAE values of 0.606 and 0.514 for different data, showing better music vocabulary recommendation. The English vocabulary personalized recommendation learning system is affirmed by most learners, more than 80% of the learners are satisfied with the reasonableness of its English vocabulary recommendation, and more than 60% of the learners think that it can promote the interest and motivation of learning English vocabulary. The development of English vocabulary personalized recommendation learning system is of practical significance to further promote the improvement of English vocabulary learning.
