Large-Scale Classroom-Based Social Recommendation Model Using Heterogeneous Graph Neural Networks and Social Regularisation
DOI:
https://doi.org/10.61360/BoniCETRAIRC262020040102Keywords:
social recommendations, heterogeneous graph neural network, social regularization, large-scale online learning, metapath, trust networkAbstract
To address the issues of information overload and learner isolation in MOOCs, this paper proposes a recommendation model (HIN-SR) based on heterogeneous graph neural networks and social regularisation. The model constructs a heterogeneous information network encompassing various entities such as learners, courses and knowledge points, and captures higher-order relationships through pedagogical semantic subpaths. Its core innovation lies in distinguishing between structural and cognitive social relationships, whilst introducing dual social regularisation: on the one hand, constraining feature similarity among friends based on trust relationships; on the other hand, measuring a learner’s influence according to their network centrality, thereby weighting their social contributions. Experiments on two real-world datasets, XuetangX and MOOCCube, demonstrate that HIN-SR significantly outperforms existing state-of-the-art models on key recommendation metrics.
References
Ahmad, I., Singh, P. V., & Gore, M. M. (2026). NGCF-RVFL: Next generation convolutional feature with random vector functional link for multi-grade diabetic retinopathy detection. Computers and Electrical Engineering, 131, 110972.
Ammar, K., Inoubli, W., Zghal, S., et al. (2025). Navigating complexity: A comprehensive review of heterogeneous information networks and embedding techniques. Knowledge and Information Systems, 67(6), 1–34. https://doi.org/10.1007/S10115-025-02357-X
Anonymous. (2009). Nevada Institute for Renewable Energy Commercialization; NIREC announces its call for proposals to enable the commercialization of renewable energy technologies. Energy Weekly News.
Avesani, G., Panico, C., Nougaret, S., et al. (2024). ESR essentials: Characterisation and staging of adnexal masses with MRI and CT—Practice recommendations by ESUR. European Radiology, 34(12), 7673–7689.
Baarir, F. N., Bourekkache, S., Ammari, A., et al. (2025). LSTM-based prediction of emotional patterns in MOOCs from behavioral data. Education and Information Technologies, 31(3), 1–28.
Camilleri, P., Watted, A., & Tonna, A. M. (2025). Investigating teachers’ changing perceptions towards MOOCs through the technology acceptance model. Education Sciences, 15(10), 1395.
Cen, Y., Jiang, S., Cai, W., et al. (2025). EGRec: A MOOCs course recommendation model based on knowledge graphs. Discover Applied Sciences, 7(6), 578.
Chen, G., He, Y., & Kong, Y. (2025). Research on recommendation algorithm based on TrustSVD++ and XGBoost. Journal of Electronic Research and Application, 9(3), 342–349.
Chen, Z., Shi, X., Zhang, S., et al. (2026). CFG-DiffNet: Canonical-face-guided diffusion feature generation for robust facial recognition under adverse conditions. Expert Systems with Applications, 307, 131035.
Chen, Z. (2025). Multimodal deep learning for sports teacher behavior analysis: Design and evaluation of a personalized continuing education recommendation system. Scientific Reports, 16(1), 1738.
Deming, S., Jingling, Y., Qing, X., et al. (2022). ACMF: An attention collaborative extended matrix factorization based model for MOOC course service via a heterogeneous view. Future Generation Computer Systems, 126, 211–224.
Farhadi, Z., Rezaei, E., Bazrafkan, L., et al. (2024). Need assessment of medical school curriculum for MOOCs: Perspectives of instructors and students of Shiraz University of Medical Sciences. BMC Medical Education, 24, 141.
Firmin, R., Schiorring, E., & Whitmer, J. (2014). Case study: Using MOOCs for conventional college coursework. Distance Education, 35(2), 178–201.
Fu, Z. (2026). Biochip-simulated genotype signals enable accurate and interpretable AMR prediction via machine learning. Frontiers in Medicine, 13, 1764292.
He, X., Deng, K., Wang, X., et al. (2020). LightGCN. In Proceedings of the International Conference on Learning Representations.
Ignacio, D., Linda, C., & Carlos, T. (2022). MOOCs as a massive learning resource for a higher education community. Education and Information Technologies, 27(9), 21–22.
Jdidou, Y., Aammou, S., Radi, E. H., et al. (2025). GEEMOO: Balancing relevance, learning outcomes, and diversity in educational recommendation systems. Intelligent Systems with Applications, 27, 200568.
Jibing, G., Cheng, W., Zhiyong, Z., et al. (2021). Automatic generation of meta-path graph for concept recommendation in MOOCs. Electronics, 10(14), 1671.
Kim, D., Jung, H., Lee, J., et al. (2026). From vulnerability to robustness: Radiation-hard isolation for CFETs. Nuclear Engineering and Technology, 58(5), 104104.
Le, V. H., & Ho, N. T. T. (2026). Balancing act: Exploring integration of Coursera MOOCs in Vietnamese higher education. Interactive Learning Environments, 34(2), 590–614.
Li, W., Wei, W., Wang, P., et al. (2025). Research on GNNs with stable learning. Scientific Reports, 15, 29002.
Liang, Y., Liu, H., & Liu, S. (2026). GDNN: A hybrid book recommendation system for ideological and political education. Electronics, 15(5), 1086.
Libin, Y., Yu, Z., Xiaoyan, C., et al. (2018). LSTM-based model for personalized context-aware citation recommendation. IEEE Access, 6, 59618–59627.
Lin, D., Liu, Y., & Wu, Z. (2026). Cross-course knowledge tracing for student performance prediction in MOOCs. Electronics, 15(3), 642.
Liu, H., Yu, J., Zhang, L., et al. (2022). NeuMF: Predicting anti-cancer drug response through a neural matrix factorization model. Current Bioinformatics, 17(9), 835–847.
Najafabadi, K. M., Kritharides, K., Choi, C., et al. (2026). Bridging the implementation gap in AI-powered personalized education: A systematic review. AI, 7(2), 41.
Peng, D. (2024). Civic education reform based on deep reinforcement learning model. Applied Mathematics and Nonlinear Sciences, 9(1).
Prabhakar, T., Prasad, M., Kumar, G., et al. (2024). High prevalence of MAFLD in general population. Alimentary Pharmacology & Therapeutics, 59(7), 843–851.
Rafailidis, D., & Daras, P. (2013). The TFC model: Tensor factorization and tag clustering for item recommendation in social tagging systems. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 43(3), 673–688.
Rodrigo, C., Dos, P. R. S., & Jonice, O. (2022). Providing recommendations for communities of learners in MOOCs ecosystems. Expert Systems with Applications, 205.
Seah, B., Bhowmick, S. S., & Dewey, F. C. (2014). DiffNet: Automatic differential functional summarization. Methods, 69(3), 247–256.
Seo, Y., & Han, S. (2010). Local scalar trust metrics with a fuzzy adjustment method. KSII Transactions on Internet and Information Systems, 4(2), 138–153.
Shaikh, M. I., Arijo, S., Noordin, K., et al. (2026). MOOCs acceptance amongst Islamic digital social finance practitioners. Information Discovery and Delivery, 54(2), 173–183.
Shangguan, A., Feng, N., Hei, X., et al. (2025). Predicting road traffic accident severity using VAE attention and GCN. Scientific Reports, 15, 34372.
Shi, C., Hu, B., Zhao, X. W., et al. (2019). Heterogeneous information network embedding for recommendation. IEEE Transactions on Knowledge and Data Engineering, 31(2), 357–370.
Shuang, L., Junlei, D., & Shu, Y. (2023). Diversified resource access paths in MOOCs. Computers & Education, 204.
Simone, B., Marco, M., Sara, L., et al. (2014). Effects of surface topography on ground shaking prediction. Geophysical Journal International, 197(3), 1551–1565.
Sun, S., Erik, S., Yang, C., et al. (2022). Mapping obesity problems scale to SF-6D. The European Journal of Health Economics, 24(2), 279–292.
Wang, T., & Ge, D. (2025). Recommendation system of online Chinese learning resources. International Journal of Human–Computer Interaction, 41(3), 1771–1781.
Wang, X., He, X., Chen, Y., et al. (2019). KGAT: Knowledge graph attention network for recommendation. arXiv preprint arXiv:1905.07854.
Wei, Z., Xue, Z., Wang, C., et al. (2026). Multi-view multi-scale dynamic graph attention network for Parkinson’s disease prediction. Scientific Reports.
Xiaona, X., & Wanxue, Q. (2023). MOOCs learning behavior propagation strategy. Education and Information Technologies, 28(10), 13349–13377.
Xu, C., Zhu, et al. (2022). Dropout prediction in XuetangX MOOCs. Neural Processing Letters, 54(4), 1–16.
Yang, Z., & Qiang, M. (2022). MP-BERT4REC: Citation recommendation. IEICE Transactions on Information and Systems, E105.D(11), 1957–1968.
Yu, X., Mao, Q., Wang, X., et al. (2024). CR-LCRP: Course recommendation model. Expert Systems with Applications, 249, 123777.
Yuan, L., Isabelle, M., & Martin, G. (2023). Social learning in crickets. Learning & Behavior, 52(3), 1–2.
Yuanfeng, L., & Yongqi, Z. (2025). Cross-domain recommendation in MOOCs. Computers and Education: Artificial Intelligence, 9, 100454.
Zeide, E., & Nissenbaum, H. (2018). Learner privacy in MOOCs. Theory and Research in Education, 16(3), 280–307.
Zhang, L., Ma, M., Zhang, J., et al. (2026). Meta-path sampling-enhanced course recommendation. Information Processing & Management, 63(3), 104482.
Zhou, Y., Li, Q., Li, R., et al. (2026). Knowledge-guided course recommendation in MOOCs. Applied Soft Computing, 193, 114768.
Zibo, L., Lan, M., & Jie, C., et al. (2022). Graph path fusion and reinforcement reasoning for recommendation in MOOCs. Education and Information Technologies, 28(1), 525–545.
Published
Issue
Section
License
Copyright (c) 2026 Jin Lu*, Aiqun Zhu
This work is licensed under a Creative Commons Attribution 4.0 International License.
