Vol. 7 No. 1 (2026)

Standard Journal Issues

Explainable Multimodal Student Profiling and Personalized Course Recommendation using Attention-Enhanced Heterogeneous Graph Neural Networks

Published 2026-01-09

  • Aamer Hashmi Syed
    • ,
  • Yashpal Singh
    • ,
  • Harshit Bhardwaj

Aamer Hashmi Syed
Department of CSE, ASET, Amity University, Rajasthan, India

Yashpal Singh
Department of CSE, ASET, Amity University, Rajasthan, India

Harshit Bhardwaj
Department of CSE, ASET, Amity University, Noida, Uttar Pradesh, India

Abstract

The rapid expansion of digital learning environments has generated rich and diverse student data, yet many existing academic support systems still rely on unimodal predictors that overlook the relational nature of learning. This work introduces an Attention-Enhanced Heterogeneous Graph Neural Network (HGNN) that unifies multimodal student profiling and personalized course recommendation within a single explainable framework. The educational ecosystem is modelled as a heterogeneous graph composed of students, courses, and socio-academic attributes, where distinct relational edges capture course enrolment patterns, peer similarity, and contextual demographic influences. An edge-type-aware attention mechanism enables the model to selectively emphasize the most influential relationships, thereby offering transparent justification for each prediction. Using a real institutional dataset of 400 learners across multiple semesters, the proposed framework achieved 94% classification accuracy surpassing conventional machine learning baselines and homogeneous graph models. The analysis of the attention revealed valuable academic and behavioral variables, which create the learning trajectory of each student. This research takes the field a step closer to trustworthy, context-sensitive, and practical educational analytics by combining classification, recommendation, and interpretability into one pipeline. It prepares the ground for early intervention, improved decision-making, and academic guidance on an individual scale.

Keywords

Heterogeneous Graph Neural Networks (HGNNs), Attention Mechanism, Multimodal Student Profiling, Intelligent Course Recommendation, Explainable AI

PDF

References

  1. A. Shahiri, W. Husain, and N. A. Rashid, “A review on predicting student’s performance using data mining techniques,” Procedia Comput. Sci., vol. 72, pp. 414–422, 2015.
  2. R. Kumar, S. K. Singh, and R. K. Aggarwal, “Student performance prediction with optimized feature selection,” Procedia Comput. Sci., vol. 172, pp. 689–695, 2020.
  3. M. Li, X. Wang, Y. Wang, Y. Chen, and Y. Chen, “Study-GNN: A novel pipeline for student performance prediction based on multi-topology graph neural networks,” Sustainability, vol. 14, no. 13, Art. 7965, Jun. 2022.
  4. Q. Hu and H. Rangwala, “Academic performance estimation with attention-based graph convolutional networks,” arXiv preprint arXiv:2001.00632, Jan. 2020.
  5. Y. Wang, A. Ding, K. Guan, S. Wu, and Y. Du, “Graph-based ensemble machine learning for student performance prediction,” arXiv preprint arXiv:2112.07893, Dec. 2021.
  6. L. Xiaochen, W. Wentao, Z. Zheng, and X. Bai, “Predicting student performance with graph structure density-based graph neural networks,” in Proc. Int. Conf. Mach. Learn. Intell. Comput. (ICMLIC), PMLR, vol. 245, pp. 25–32, 2024.
  7. X. Zhang, Y. Zhang, A. L. Chen, M. Yu, and L. Zhang, “Optimizing multi-label student performance prediction with GNN-TINet: A contextual multidimensional deep learning framework,” PLoS One, vol. 20, no. 1, e0314823, Jan. 2025.
  8. T. H. Nguyen, T. B. Nguyen, and H. V. Nguyen, “Multi-view hypergraph neural networks for student academic performance prediction,” Expert Syst. Appl., vol. 215, 119514, 2023.
  9. S. X. Zhang, J. Bai, and Q. Wang, “Improving academic performance predictions with dual graph neural networks,” Appl. Intell., vol. 54, pp. 15879–15894, 2024.
  10. Y. Ektefaie, G. Dasoulas, A. Noori, M. Farhat, and M. Zitnik, “Multimodal learning with graphs,” Nat. Mach. Intell., vol. 5, no. 4, pp. 345–357, Apr. 2023.
  11. J. Yuan and K. Wang, “A survey on graph-based educational analytics,” Comput. Educ., vol. 207, 104401, 2023.
  12. F. Qian, L. Bai, L. Cui, M. Li, and H. Du, “Exploring the over-smoothing problem of graph neural networks for graph classification: An entropy-based viewpoint,” in Proc. AAAI Conf. Artif. Intell., 2025, pp. 19995–20003.
  13. T. Cai, Y. Jiang, M. Li, and C. Huang, “ML-GOOD: Multi-label graph out-of-distribution detection,” in Proc. AAAI Conf. Artif. Intell., 2025, pp. 15650–15658.
  14. P. Hooda, P. Mittal, B. D. Veerasamy, R. Bhatt, C. S. Lakshmi, S. Kamal, and P. K. Shukla, "Enhancing churn prediction through advanced machine learning techniques for modern education in computer science," Int. J. Mod. Educ. Comput. Sci. (IJMECS), vol. 17, no. 1, pp. 91–101, 2025, doi: 10.5815/ijmecs.2025.01.07.
  15. L. Bai, L. Cui, Z. Zhao, and M. Li, “AEGK: Aligned entropic graph kernels through continuous-time quantum walks,” IEEE Trans. Knowl. Data Eng., vol. 37, no. 3, pp. 1064–1078, 2025.
  16. S. Luan, C. Hua, Q. Lu, and L. Wu, “Revisiting heterophily for graph neural networks,” in Adv. Neural Inf. Process. Syst. (NeurIPS), 2023.
  17. N. Yanes, A. M. Mostafa, M. Ezz, and S. N. Almuayqil, “A machine learning-based recommender system for improving students learning experiences,” IEEE Access, vol. 8, pp. 185577–185588, 2020, doi: 10.1109/ACCESS.2020.3036336.
  18. P. Kinnunen, R. McCartney, L. Murphy, and L. Thomas, “Through the eyes of instructors: A phenomenographic investigation of student success,” in Proc. 3rd Int. Comput. Educ. Res. Workshop (ICER’07), pp. 61–72, 2007, doi: 10.1145/1288580.1288589.
  19. V. Tam, E. Y. Lam, S. T. Fung, and W. W. T. Fok, “Enhancing educational data mining techniques on online educational resources with a semi-supervised learning approach,” in Proc. Int. Conf. Comput. Educ. (ICCE), Dec. 2015, pp. 203–206.
  20. N. A. Butt, Z. Mahmood, K. Shakeel, S. Alfarhood, M. Safran, and I. Ashraf, “Performance prediction of students in higher education using multi-model ensemble approach,” IEEE Access, vol. 11, pp. 136091–136108, 2023, doi: 10.1109/ACCESS.2023.3336987.
  21. S. A. Hashmi, Y. Singh, and H. Bhardwaj, “Improving student classification model using Boruta feature selection technique,” in Communication and Intelligent Systems, H. Sharma, V. Shrivastava, A. K. Tripathi, and L. Wang, Eds., Lecture Notes in Networks and Systems, vol. 1371. Singapore: Springer, 2025, pp. 165–176, doi: 10.1007/978-981-96-5723-0_13.
  22. Q. Al-Tashi, S. J. Abdul Kadir, H. M. Rais, S. Mirjalili, and H. Alhussian, “Binary optimization using hybrid grey wolf optimization for feature selection,” IEEE Access, vol. 7, pp. 39496–39508, 2019, doi: 10.1109/ACCESS.2019.2906757.
  23. V. Hegde and H. S. Rao, “A framework to analyze performance of students in programming language using educational data mining,” in Proc. IEEE Int. Conf. Comput. Intell. Comput. Res. (ICCIC), 2017, pp. 1–4, doi: 10.1109/ICCIC.2017.8524244.
  24. P. Rojanavasu, “Educational data analytics using association rule mining and classification,” in Proc. Joint Int. Conf. Digit. Arts, Media Technol. with ECTI Northern Sect. Conf. Electr., Electron., Comput. Telecommun. Eng. (ECTI DAMT-NCON), 2019, pp. 142–145.
  25. P. M. Arsad, N. Buniyamin, and J. L. A. Manan, “A neural network students’ performance prediction model (NNSPPM),” in Proc. IEEE Int. Conf. Smart Instrum., Meas. Appl. (ICSIMA), 2013, pp. 26–27, doi: 10.1109/ICSIMA.2013.6717966.
  26. ?. Aydo?du, “A new student modeling technique with convolutional neural networks: LearnerPrints,” J. Educ. Comput. Res., vol. 59, no. 4, pp. 603–619, 2021, doi: 10.1177/0735633120969216.
  27. K. Okoye et al., “A text mining and statistical approach for assessment of pedagogical impact of students’ evaluation of teaching and learning outcome in education,” IEEE Access, vol. 11, pp. 15348–15360, Feb. 2023, doi: 10.1109/ACCESS.2023.3239779.
  28. V. U. Kumar, A. Krishna, P. Neelakanteswara, and C. Z. Basha, “Advanced prediction of performance of a student in a university using machine learning techniques,” in Proc. Int. Conf. Electron. Sustain. Commun. Syst. (ICESC), 2020, pp. 121–126, doi: 10.1109/ICESC48915.2020.9155557.
  29. K. Al Mayahi and D. M. Al-Bahri, “Machine learning based predicting student academic success,” in Proc. Int. Congr. Ultra Modern Telecommun. Control Syst. Workshops (ICUMT), Oct. 2020, pp. 264–268, doi: 10.1109/ICUMT51630.2020.9222435.
  30. E. G. Rincon-Flores, E. Lopez-Camacho, J. Mena, and O. O. Lopez, “Predicting academic performance with artificial intelligence (AI), a new tool for teachers and students,” in Proc. IEEE Global Eng. Educ. Conf. (EDUCON), Apr. 2020, pp. 1049–1054, doi: 10.1109/EDUCON45650.2020.9125141.
  31. A. Sharma and S. Kumar, “An improved student performance prediction model using ensemble learning,” IEEE Access, vol. 10, pp. 105–115, 2022.
  32. M. Chen, Y. Zhang, L. Wang, and K. Huang, “Student academic risk prediction using XGBoost and feature selection,” Applied Sciences, vol. 11, no. 23, pp. 11234–11247, Dec. 2021.
  33. D. Monteverde-Suárez, R. Garrote, and M. Ferrando, “Predicting students’ academic progress and related attributes in first-year medical students: an analysis with ANN and Naïve Bayes,” BMC Medical Education, vol. 24, no. 74, pp. 1–14, 2024.
  34. S. Kumar and F. Janan, “Prediction of student’s performance using Random Forest classifier,” International Journal of Computer Applications, vol. 183, no. 15, pp. 25–30, 2021.

Downloads

Download data is not yet available.

Similar Articles

11-20 of 63

You may also start an advanced similarity search for this article.