In vocational education, learners come from diverse backgrounds, possess various skills, and have different preferences, making personalized instruction essential for effective skill development. Traditional teaching approaches often lack adaptability to individual needs, resulting in disengagement and limited learning outcomes. This research aims to develop a User-Based Collaborative Filtering (UBCF) algorithm to deliver personalized course and resource recommendations using comprehensive vocational education assessment data. The model integrates various data sources, including student demographic and academic profiles, practical and theoretical skill assessment scores, learning preferences, behavior logs, course and content metadata, and historical performance with feedback. Data preprocessing involves handling missing values and Z-score normalization to ensure data quality. Term frequency-inverse document frequency (TF-IDF) was applied for feature extraction from textual content and feedback. Feature selection was optimized using Dynamic Glowworm Swarm Optimization (DGSO), ensuring relevant and non-redundant attributes were selected to enhance recommendation accuracy and computational efficiency. A UBCF model was implemented to identify similar learners based on their profiles and behaviors, enabling the system to recommend relevant learning resources. A feedback loop was included to refine the model’s accuracy and effectiveness iteratively. Experimental evaluation using real-world vocational education datasets in Python demonstrated that the proposed model effectively generated personalized and diverse course recommendations. The system significantly improved in accuracy (96%) and MAE (0.04), with individual learning needs and preferences. The proposed UBCF-based recommendation system successfully enhances the personalization of vocational education. By leveraging multi-source data and intelligent filtering, it contributes to adaptive learning environments that support better learner performance and satisfaction.