Abstract

A key challenge in higher education is providing personalized course selections tailored to individuals’ varying levels of academic knowledge, cognitive skills, and learning competencies. Traditional methods primarily focus on historical preferences, similarities, or interactions without considering such nuanced differences. In this study, Studyproposes an expert-guided cognitive-competency machine learning framework that enables explainable personalized course recommendation through integration of both student-specific cognitive skill profiles and domain expertise. Studybuilds upon the premise that successful course recommendation should reflect how well learners' cognitive capabilities align with various subjects, which can only be fully realized when combining cognitive/academic attributes with expertly defined skill requirements for courses/specializations. Studytransforms these heterogeneous attributes into an integrated representation and use multiple supervised machine learning models to learn the relationships between learner profiles and course domains deemed suitable. Subsequently, a ranking process applies to derive recommendations based on predicted relevance scores. Finally, recommenders incorporate mechanisms to provide explanations regarding why particular cognitive and competency aspects contribute to specific recommendations. To evaluate the framework’s effectiveness, experiments consider various dimensions ranging from comparative model evaluations and contributions analyses to ablation studies and case-based analysis. results demonstrate the importance of incorporating diverse information sources – whether individually or together - when seeking effective academic guidance. The proposed approach thus lays a foundational step towards enabling more personalized, transparent, and competence-oriented academic advising. It also opens possibilities for extensions such as adaptive learner profiling and real-time educational recommendation systems.

References

• S. Bian, “Dynamic optimization algorithm for personalized Japanese learning path driven by reinforcement learning,” Discov. Artif. Intell., vol. 6, no. 1, Dec. 2026, doi: 10.1007/S44163-026-01979-6.
• X. Zhang and X. Chang, “Hybrid deep learning and IoT-based intelligent learning analytics system for personalized education,” Discov. Comput., vol. 29, no. 1, Dec. 2026, doi: 10.1007/S10791-026-10419-Z.
• J. Li, “Deep learning-based personalized learning path planning and optimization model for music education,” Discov. Artif. Intell., vol. 6, no. 1, Dec. 2026, doi: 10.1007/S44163-026-01115-4.
• X. Yang, J. Y. Feng, S. Y. Guo, B. X. Fang, and Q. Liao, “FedPD: personalized federated learning based on partial distillation,” Front. Comput. Sci., vol. 20, no. 3, Mar. 2026, doi: 10.1007/S11704-025-40840-4.
• D. H. Adhya and S. Panda, “An integrative systematic review analysis of research on technology-enabled personalized learning and self-regulated learning,” Smart Learn. Environ., vol. 13, no. 1, Dec. 2026, doi: 10.1186/S40561-025-00428-8.
• M. Li, X. He, and J. Chen, “Personalized federated learning with prototype representation and robust aggregation,” Cluster Comput., vol. 29, no. 2, Apr. 2026, doi: 10.1007/S10586-025-05850-8.
• Z. Fu, “Integrating Reinforcement Learning with Dynamic Knowledge Tracing for personalized learning path optimization,” Sci. Rep., vol. 15, no. 1, Dec. 2025, doi: 10.1038/S41598-025-23900-4.
• T. S. S. Kumar and R. Thandeeswaran, “A rule-based personalized educational learning classifier system for programming instruction,” Knowl. Inf. Syst., vol. 68, no. 1, Dec. 2026, doi: 10.1007/S10115-026-02775-5.
• J. Jiang, J. Huang, H. Ren, Y. Long, and H. Sang, “LLM-CR: LLM-enhanced course recommendation,” Int. J. Educ. Technol. High. Educ., vol. 23, no. 1, Dec. 2026, doi: 10.1186/S41239-026-00590-0.
• Z. Li, M. Xie, Y. Zeng, Q. Lei, and R. Li, “RHCR: a reinforced heterogeneous knowledge graph for course recommendation,” Sci. Rep., Jun. 2026, doi: 10.1038/S41598-026-56490-W.
• W. Yao and X. Hu, “Large Language Model-empowered Course Recommendation with Learning Interest-Goal Contrastive Learning,” J. Syst. Sci. Syst. Eng., 2026, doi: 10.1007/S11518-025-5702-8.
• J. Qian and H. Fan, “Deep knowledge tracking and course recommendation model for online education learners,” Discov. Artif. Intell., vol. 6, no. 1, Dec. 2026, doi: 10.1007/S44163-026-01977-8.
• N. Adhvaryu and A. Dave, “Implementation of recommendation system using content based approach for course selection of university students,” SN Bus. Econ., vol. 6, no. 6, Jun. 2026, doi: 10.1007/S43546-026-01156-Y.
• W. Fu, “Learning Like a Human: Zero-shot Personalized Course Recommendation System via Cognitive Diagnosis and LLM Agents,” Int. J. Comput. Intell. Syst., May 2026, doi: 10.1007/S44196-026-01354-4.
• W. Deng, H. Chen, L. Lu, P. Zhu, and J. Wu, “A hierarchical attention neural network with multi-view fusion for online course recommendation,” Knowl. Inf. Syst., vol. 67, no. 8, pp. 7057–7077, Aug. 2025, doi: 10.1007/S10115-025-02441-2.
• T. K. F. Chiu, M. Çoban, I. T. Sanusi, and M. A. Ayanwale, “Validating student AI competency self-efficacy (SAICS) scale and its framework,” Educ. Technol. Res. Dev., vol. 73, no. 4, pp. 2785–2807, Aug. 2025, doi: 10.1007/S11423-025-10512-Y.
• S. Armandpishe, M. Salehpoor-Emran, M. Rahimzadeh, S. Saeidi, F. Jafari, and M. Banazadeh, “Association between clinical reasoning competency and learning behavior among undergraduate nursing students: a cross-sectional study,” BMC Med. Educ., vol. 26, no. 1, Dec. 2026, doi: 10.1186/S12909-026-08668-Z.
• P. Chunhabundit, T. Arayapisit, and N. Srimaneekarn, “Evaluating dental student competency in forensic rugoscopy for human identification,” Sci. Rep., vol. 16, no. 1, Dec. 2026, doi: 10.1038/S41598-025-33373-0.
• S. K. Boonstra, K. Kretke, D. Garcia, H. Fischer, and M. Storksdieck, “Technical STEM Workforce Development Re-Imagined: The Lucy Student Pipeline and Competency Enabler (L’SPACE),” Space Sci. Rev., vol. 221, no. 2, Mar. 2025, doi: 10.1007/S11214-024-01123-9.
• R. Gutierrez, A. Maldonado, and W. Villegas-Ch, “Adaptive mathematical competency assessment using graph neural networks and reinforcement learning,” Discov. Artif. Intell., vol. 6, no. 1, Dec. 2026, doi: 10.1007/S44163-026-00924-X.
• M. H. Musa, S. Salam, M. A. Norasikin, W. Hall, and J. F. Rusdi, “Advancing semantic search in higher education through the Heterogeneous Student Performance data retrieval model,” Discov. Educ., vol. 5, no. 1, Dec. 2026, doi: 10.1007/S44217-026-01580-0.