Research on Personalized Learning Resource Recommendation Algorithm for Employment and Entrepreneurship Education Curriculum Based on Data Mining

Employment and entrepreneurship courses lack unified resource descriptions, hindering recommendation algorithms. We propose a data-mining-based algorithm using the H-BIRCH for efficient resource clustering and semantic analysis. Processing with variational mode decomposition, wavelet denoising, and feature extraction improves data quality. Considering user interests, hotspots, and attribute satisfaction, personalized recommendations boost matching accuracy and learning efficiency. Experiments show semantic relevance of 0.9-0.97 and a stable user click-through rate of 57%.