Adaptive Federated Learning Framework for Fair and Privacy-Preserving Educational Analytics
DOI:
https://doi.org/10.31185/wjps.988Keywords:
Federated learning, adaptive aggregation, educational data mining, privacy preservation, fairness, label-flipping attackAbstract
The rapid digitalization of educational environments has led to the large-scale generation of student-related data from online learning platforms, intelligent tutoring systems, and virtual classrooms. While such data enable advanced predictive analytics and personalized learning, they also raise critical concerns related to data privacy, fairness, and ethical use. To address these challenges, this paper proposes an Adaptive Federated Learning (AFL) framework for fair and privacy-preserving educational analytics. Unlike conventional centralized learning approaches, the proposed framework allows multiple educational institutions to collaboratively train predictive models without sharing raw student data. Moreover, AFL incorporates a fairness-aware adaptive aggregation mechanism that dynamically adjusts client contributions based on both local predictive performance and fairness indicators. This strategy improves model robustness and reduces demographic bias under heterogeneous and potentially adversarial data distributions. The effectiveness of the proposed framework is evaluated on three publicly available educational datasets—Student Performance, Predict Students’ Dropout and Academic Success, and the Open University Learning Analytics Dataset (OULAD)—under both normal and label-flipping attack scenarios. Experimental results demonstrate that AFL achieves performance comparable to centralized models while consistently improving fairness and resilience against adversarial behavior. These findings highlight the potential of AFL as a trustworthy and ethically aligned solution for decentralized educational data analytics.
References
[1] B. Liu et al., “Recent advances on federated learning: A systematic survey,” Neurocomputing, vol. 592, Art. no. 127657, pp. 1–27, 2024, doi: 10.1016/j.neucom.2024.127657.
[2] S. W. A. Alsudani and A. Ghazikhani, “Enhancing intrusion detection with LSTM recurrent neural networks optimized by the emperor penguin algorithm,” World Journal of Computing and Machine Systems, vol. 2, no. 3, pp. 69–80, 2023, doi: 10.31185/wjcms.166. DOI: https://doi.org/10.31185/wjcms.166
[3] T. H. Rafi, M. B. Islam, and A. Anwar, “Fairness and privacy preservation in federated learning: A survey,” Future Generation Computer Systems, vol. 157, pp. 299–321, 2024, doi: 10.1016/j.future.2024.02.008. DOI: https://doi.org/10.1016/j.future.2024.02.008
[4] T. Salazar, H. Araújo, A. Cano, and P. H. Abreu, “A survey on group fairness in federated learning: Challenges, taxonomy of solutions, and future directions,” arXiv preprint, arXiv:2410.03855, 2024.
[5] N. Mukhtiar et al., “Federated learning at the forefront of fairness,” in Proc. Int. Joint Conf. Artificial Intelligence (IJCAI), 2025, pp. 10615–10623. DOI: https://doi.org/10.24963/ijcai.2025/1177
[6] N. Benarba et al., “Bias in federated learning: A comprehensive survey,” ACM Computing Surveys, vol. 57, no. 8, Art. no. 174, 2025, doi: 10.1145/3682301. DOI: https://doi.org/10.1145/3735125
[7] J. Pei et al., “F3: Fair federated learning with adaptive regularization,” Knowledge-Based Systems, vol. 317, Art. no. 111310, 2025, doi: 10.1016/j.knosys.2025.111310.
[8] J. He et al., “FedAA: A reinforcement learning perspective on adaptive aggregation in federated learning,” in Proc. AAAI Conf. Artificial Intelligence, vol. 39, no. 9, 2025, pp. 9614–9622. DOI: https://doi.org/10.1609/aaai.v39i16.33878
[9] S. Alsudani, H. Nasrawi, M. Shattawi, and A. Ghazikhani, “Enhancing spam detection using a crow-optimized FFNN with LSTM,” World Journal of Computing and Machine Systems, vol. 3, no. 1, pp. 28–39, 2024, doi: 10.31185/wjcms.199. DOI: https://doi.org/10.31185/wjcms.199
[10] J. Seo et al., “GC-Fed: Gradient centralized federated learning with partial client participation,” arXiv preprint, arXiv:2503.13180, 2025. DOI: https://doi.org/10.1016/j.inffus.2026.104148
[11] N. M. Jebreel, F. Thabtah, and S. M. Alshahrani, “LFighter: Defending against label-flipping attacks in federated learning,” Neural Networks, vol. 172, pp. 106–121, 2024, doi: 10.1016/j.neunet.2023.10.010. DOI: https://doi.org/10.2139/ssrn.5528538
[12] L. Lavaur, F. Guillemin, and P. Vicat-Blanc, “Investigating the impact of label-flipping attacks against federated learning for intrusion detection systems,” Computers & Security, vol. 140, Art. no. 104893, 2025, doi: 10.1016/j.cose.2024.104893. DOI: https://doi.org/10.1016/j.cose.2025.104462
[13] M. Alshawabkeh et al., “Systematic analysis of label-flipping attacks against federated learning for intrusion detection systems,” in Proc. ACM Asia Conf. Computer and Communications Security (AsiaCCS), 2024, pp. 1615–1627, doi: 10.1145/3658644.3658718.
[14] W. Ma et al., “A defense method against multi-label poisoning attacks in federated learning,” Scientific Reports, vol. 15, no. 1, Art. no. 10514, 2025, doi: 10.1038/s41598-025-10514-6. DOI: https://doi.org/10.1038/s41598-025-09672-x
[15] S. Alsudani and M. N. Saeea, “Enhancing thyroid disease diagnosis using the emperor penguin optimization algorithm,” Wasit Journal for Pure Sciences, vol. 2, no. 4, pp. 66–79, 2023, doi: 10.31185/wjps.230. DOI: https://doi.org/10.31185/wjps.230
[16] S. Chen et al., “Privacy-preserving federated learning via homomorphic encryption,” arXiv preprint, arXiv:2412.01650, 2024.
[17] B. Zhu et al., “A privacy-preserving federated learning scheme with homomorphic encryption and secure aggregation,” Ain Shams Engineering Journal, vol. 16, no. 3, Art. no. 102274, 2025, doi: 10.1016/j.asej.2024.102274.
[18] L. M. Lopez-Ramos, F. Guillemin, and P. Vicat-Blanc, “Interplay between federated learning and explainable AI: Opportunities, challenges, and threats,” arXiv preprint, arXiv:2411.05874, 2024.
[19] R. Kalakoti et al., “Federated learning of explainable AI for deep intrusion detection,” Computer Networks, vol. 252, Art. no. 110679, 2025, doi: 10.1016/j.comnet.2025.110679. DOI: https://doi.org/10.2139/ssrn.5043292
[20] P. Dubey et al., “Integrating explainable AI with federated learning for next-generation IoT: A survey,” Computer Communications, vol. 228, pp. 79–101, 2025, doi: 10.1016/j.comcom.2025.01.011. DOI: https://doi.org/10.1016/j.cosrev.2024.100697
[21] K. Bonawitz et al., “Towards federated learning at scale: System design,” arXiv preprint, arXiv:1902.01046, 2019.
[22] M. Ekmefjord et al., “Scalable federated machine learning with FEDn,” arXiv preprint, arXiv:2103.00148, 2021.
[23] A. Grataloup et al., “A systematic survey on the application of federated learning to mental health and human activity recognition,” Frontiers in Digital Health, vol. 6, Art. no. 1426983, 2024, doi: 10.3389/fdgth.2024.1426983. DOI: https://doi.org/10.3389/fdgth.2024.1495999
[24] T. Zhang et al., “Enhancing dropout prediction in distributed educational platforms using federated learning,” Mathematics, vol. 11, no. 24, Art. no. 4977, 2023, doi: 10.3390/math11244977. DOI: https://doi.org/10.3390/math11244977
[25] M. Quimiz-Moreira, P. Segura-Fernández, and D. García-González, “Factors, prediction, explainability, and simulating interventions for college dropout: A literature review (2012–2024),” Computers, vol. 13, no. 8, Art. no. 198, 2025, doi: 10.3390/computers13080198. DOI: https://doi.org/10.3390/computation13080198
[26] C. Chen et al., “Enforcing group fairness in privacy-preserving federated learning,” Future Generation Computer Systems, vol. 160, pp. 890–900, 2024. DOI: https://doi.org/10.1016/j.future.2024.06.040
[27] A. A. Bendoukha et al., “Towards privacy-preserving and fairness-aware federated learning frameworks,” Proceedings on Privacy Enhancing Technologies, vol. 2025, no. 1, pp. 845–865, 2025. DOI: https://doi.org/10.56553/popets-2025-0044
[28] H. B. McMahan et al., “Communication-efficient learning of deep networks from decentralized data,” in Proc. Int. Conf. Artificial Intelligence and Statistics (AISTATS), Fort Lauderdale, FL, USA, 2017, pp. 1273–1282.
[29] B. Baruch, M. Baruch, and Y. Goldberg, “A little is enough: Circumventing defenses for distributed learning,” in Proc. Advances in Neural Information Processing Systems (NeurIPS), 2019, pp. 8635–8645.
[30] X. Fang, M. Ye, and P. Cheng, “Robust federated learning against data poisoning attacks,” IEEE Transactions on Neural Networks and Learning Systems, vol. 34, no. 1, pp. 460–472, 2023, doi: 10.1109/TNNLS.2021.3137938.
[31] S. W. A. Alsudani and G. K. Saud, “Recurrent neural networks optimized by grasshopper algorithm for audio-based Parkinson’s disease diagnosis,” Wasit Journal for Pure Sciences, vol. 4, no. 2, pp. 56–75, 2025, doi: 10.31185/wjps.766. DOI: https://doi.org/10.31185/wjps.766
[32] S. Barocas, M. Hardt, and A. Narayanan, Fairness and Machine Learning: Limitations and Opportunities. Cambridge, MA, USA: MIT Press, 2019.
[33] S. L. Cheng and H. Wang, “FedGCR: Achieving performance and fairness for federated learning with distinct client types via group customization and reweighting,” in Proc. AAAI Conf. Artificial Intelligence, vol. 38, no. 10, 2024, pp. 11023–11031. DOI: https://doi.org/10.1609/aaai.v38i10.29031
[34] M. Badar and S. Sikdar, “FairTrade: Achieving Pareto-optimal trade-offs between balanced accuracy and fairness in federated learning,” in Proc. AAAI Conf. Artificial Intelligence, vol. 38, no. 9, 2024, pp. 10012–10020. DOI: https://doi.org/10.1609/aaai.v38i10.28971
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Hussein Nasrawi
This work is licensed under a Creative Commons Attribution 4.0 International License.
