An Experimental Framework for Student Course Recommendation Using Large Language Models

Year 2025, Volume: 1 Issue: 1, 21 - 28, 31.05.2025

Alican Doğan

Abstract

This study presents a novel approach to student course recommendation by leveraging Large Language Models (LLMs), specifically ChatGPT, as an intelligent assistant in academic advising. Traditional recommendation systems in educational settings often rely on fixed rules or collaborative filtering techniques, which may fail to capture the nuanced goals and interests of individual students. In contrast, LLMs can interpret natural language inputs, allowing for more personalized and context-aware suggestions. Within this framework, we simulate student profiles based on academic history, interests, and career goals, and evaluate the quality of course recommendations generated by the model. Preliminary results show that LLMs can provide coherent, relevant, and goal-aligned suggestions without the need for extensive training data or domain-specific tuning. This work highlights the potential of using general-purpose language models to support student advisory services, especially in institutions where counselor resources are limited. The proposed framework can serve as a foundation for future AI-integrated educational support systems.

Keywords

LLM , ChatGPT , Course Recommendation , Student Support , Artificial Intelligence in Education

Büyük Dil Modelleri Kullanarak Öğrenci Ders Tavsiyesi için Deneysel Bir Çerçeve

Abstract

Bu çalışma, öğrencilere ders önerisi sunmada ChatGPT gibi Büyük Dil Modelleri'ni (LLM'ler) akademik danışmanlıkta akıllı bir asistan olarak kullanarak yenilikçi bir yaklaşım ortaya koymaktadır. Eğitim ortamlarında kullanılan geleneksel öneri sistemleri genellikle sabit kurallara veya iş birliğine dayalı filtreleme tekniklerine dayanmakta olup, öğrencilerin bireysel hedef ve ilgi alanlarını yeterince yansıtamamaktadır. Buna karşın, LLM'ler doğal dil girdilerini yorumlayabildikleri için daha kişiselleştirilmiş ve bağlamsal öneriler sunabilmektedir. Bu çerçevede, öğrencilerin akademik geçmişi, ilgi alanları ve kariyer hedeflerine dayalı olarak öğrenci profilleri simüle edilmiş ve modelin oluşturduğu ders önerilerinin kalitesi değerlendirilmiştir. İlk sonuçlar, LLM’lerin kapsamlı eğitim verilerine veya alana özgü ayarlamalara ihtiyaç duymadan tutarlı, ilgili ve hedefe uygun öneriler sunabildiğini göstermektedir. Bu çalışma, özellikle danışmanlık kaynaklarının kısıtlı olduğu kurumlarda, genel amaçlı dil modellerinin öğrenci rehberlik hizmetlerini desteklemede sahip olduğu potansiyele dikkat çekmektedir. Önerilen çerçeve, gelecekte yapay zekâ entegreli eğitim destek sistemleri için bir temel teşkil edebilir.

Keywords

LLM , ChatGPT , Ders Önerisi , Öğrenci Desteği , Eğitimde Yapay Zeka

References

• A. Goslen, Y. J. Kim, J. Rowe, and J. Lester. (2024). "LLM-based Student Plan Generation for Adaptive Scaffolding in Game-Based Learning Environments" International Journal of Artificial Intelligence in Education, vol. 34, pp. 1–26. doi:10.1007/s40593-024-00421-1.
• O. Henkel, L. Hills, B. Roberts, and J. McGrane. (2024). "Can LLMs Grade Open Response Reading Comprehension Questions? An Empirical Study Using the ROARs Dataset" International Journal of Artificial Intelligence in Education, vol. 34, pp. 1–26. doi:10.1007/s40593-024-00431-z.
• C. Li and W. Xing. (2021). "Natural Language Generation Using Deep Learning to Support MOOC Learners" International Journal of Artificial Intelligence in Education, vol. 31, pp. 186–214. doi: 10.1007/s40593-020-00235-x.
• J. Lin, Z. Han, D. R. Thomas, A. Gurung, S. Gupta, V. Aleven, and K. R. Koedinger. (2024). "How Can I Get It Right? Using GPT to Rephrase Incorrect Trainee Responses" International Journal of Artificial Intelligence in Education, vol. 34, pp. 1–27. doi:10.48550/arXiv.2405.00970.
• W. Morris, L. Holmes, J. S. Choi, and S. Crossley. (2024). "Automated Scoring of Constructed Response Items in Math Assessment Using Large Language Models," International Journal of Artificial Intelligence in Education, vol. 34, pp. 1–28. doi: 10.1007/s40593-024-00418-w.
• K. A. Norberg, H. Almoubayyed, L. De Ley, A. Murphy, K. Weldon, and S. Ritter. (2024) "Rewriting Content with GPT-4 to Support Emerging Readers in Adaptive Mathematics Software" International Journal of Artificial Intelligence in Education, vol. 34, pp. 1–40. doi:10.1007/s40593-024-00420-2.
• M. J. Parker, C. Anderson, C. Stone, and Y. Oh. (2024). "A Large Language Model Approach to Educational Survey Feedback Analysis" International Journal of Artificial Intelligence in Education, vol. 34, pp. 1–38. doi:10.1007/s40593-024-00414-0
• S. Sarsa, P. Denny, A. Hellas, and J. Leinonen. (2022). "Automatic Generation of Programming Exercises and Code Explanations Using Large Language Models" in Proceedings of the 2022 ACM Conference on International Computing Education Research, pp. 27–43. doi:10.48550/arXiv.2206.11861.
• Z. Wang, A. Lan, and R. Baraniuk. (2021)."Math Word Problem Generation with Mathematical Consistency and Problem Context Constraints" in Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pp. 5986–5999. doi: 10.18653/v1/2021.emnlp-main.484.
• F. Zhang, C. Li, O. Henkel, W. Xing, S. Baral, L. Hefernan, and H. Li. (2024). "Math-LLMs: AI Cyberinfrastructure with Pre-trained Transformers for Math Education" International Journal of Artificial Intelligence in Education, vol. 34, pp. 1–24. doi:10.1007/s40593-024-00416-y.
• G. Kismetova and M. Amirhanova. (2024). "Artificial Intelligence in Teaching English: Prospects and Limitations" Slovak international scientific journal, vol. 89, pp. 34–37. doi:10.5281/zenodo.14167040.
• N. Anantrasirichai and D. Bull. (2021). "Artificial Intelligence in the Creative Industries: A Review" Artificial Intelligence Review, vol. 55, pp. 589–656. doi: 10.1007/s10462-021-10039-7.
• K. Lekan and Z. Pardos. (2025). "AI-Augmented Advising," Journal of Learning Analytics, vol. 12, no. 1, pp. 110–128. doi: 10.18608/jla.2025.8593.
• H. Van Deventer, M. Mills, and A. Evrard. (2024). "From Interests to Insights: An LLM Approach to Course Recommendations Using Natural Language Queries" arXiv preprint arXiv:2412.19312. doi:10.48550/arXiv.2412.19312.
• A. Z. Khan and A. Polyzou. (2023). "Session-based Course Recommendation Frameworks Using Deep Learning" Proceedings of the 16th International Conference on Educational Data Mining, pp. 269–277. doi:10.5281/zenodo.8115699.
• H. Nguyen, T. Tran, and L. Pham. (2024). "Extracting Course Features and Learner Profiling for Course Recommendation," International Review of Research in Open and Distributed Learning, vol. 25, no.1, pp. 197–225. doi:10.19173/irrodl.v25i1.7419.