Federated and Privacy-Preserving Adaptive Learning Framework for AI-Driven Mobile Education Platforms

D. Saisanthiya; G. Malarselvi; G. Jessy Sujana; V. Saidulu; K. Rajesh Kumar

doi:10.3991/ijim.v20i12.62166

Authors

D. Saisanthiya SRM Institute of Science and Technology, Chengalpattu, India https://orcid.org/0009-0008-8494-790X
G. Malarselvi SRM Institute of Science and Technology, Chengalpattu, India
G. Jessy Sujana SRM Institute of Science and Technology, Chennai, India https://orcid.org/0009-0005-4568-229X
V. Saidulu JNTUH University, Hyderabad, India
K. Rajesh Kumar SRM Institute of Science and Technology, Tiruchirappalli, India https://orcid.org/0000-0003-0411-020X

DOI:

https://doi.org/10.3991/ijim.v20i12.62166

Keywords:

Federated Learning, Differential Privacy, Adaptive Learning, Mobile Education, Privacy-Preserving AI, Non-IID Data, Secure Aggregation, Personalized Learning

Abstract

AI-enabled mobile learning platforms have experienced explosive growth, enabling personalised and scalable educational opportunities, but the centralisation of learner data through collection and processing raises serious privacy, security, and regulatory issues. Existing adaptive learning solutions use central architectures that put sensitive student data at risk and do not facilitate inter-institutional cooperation. To tackle these challenges, we propose a Federated and Privacy-Preserving Adaptive Learning framework (FedPAL) to facilitate adaptive learning in intelligent mobile education environments. In this respectful paper, we propose a framework that integrates three pivotal aspects: (i) an Adaptive Engine based on Federated Learning (FLAE), facilitating the training of models in partnership with distributed mobile devices or educational institutions without data sharing on learner raw data; (ii) Privacy-Preserving Optimisation (PPO) using Differential Privacy and Secure Aggregation to ensure user-level information remains obscured while sustaining model performance metrics; and finally, (iii) Personalised Learning Orchestration Layer (PLOL), flexibly adapting content dissemination according to learner behaviour, performance indicators, and contextual usage patterns. To tackle the heterogeneous and non-IID learning data owned by users, an Adaptive Client Weighting Mechanism (ACWM) is proposed. Real-time responsiveness and low latency are guaranteed by lightweight on-device inference models. Experimental evaluation on multi-institutional mobile learning datasets shows that the accuracy (ACC) of our approach increases by 3.8–6.5% compared to centralised and baseline federated approaches, while maintaining strong privacy guarantees (ε ≤ 5, δ = 10−5). FedPALs high performance under partial participation and network heterogeneity makes it suitable for practical deployment.

References

[1] Zawacki-Richter, O., Marín, V. I., Bond, M., & Gouverneur, F. (2019). Systematic review of research on artificial intelligence applications in higher education—where are the educators. of Educational Technology in Higher Education, 16(1), 39. https://doi.org/10.1186/s41239-019-0171-0

[2] Luckin, R., Holmes, W., Griffiths, M., & Forcier, L. B. (2016). Intelligence unleashed: An argument for AI in education. Pearson Education. ISBN: 978-0-992-56533-5.

[3] Choi, Y., Lee, Y., Cho, J., Baek, J., Kim, B., Cha, Y., ... & Heo, J. (2020). Towards an appropriate query, key, and value computation for knowledge tracing. Proceedings of the 7th ACM Conference on Learning @ Scale (L@S), 160–169. https://doi.org/10.1145/3386527.3405945

[4] VanLehn, K. (2011). The relative effectiveness of human tutoring, intelligent tutoring systems, and other tutoring systems. Educational Psychologist, 46(4), 197–221. https://doi.org/10.1080/00461520.2011.611369

[5] Williamson, B. (2019). Policy networks, performance metrics and platform markets: Charting the expanding data infrastructure of K–12 educational governance in England. British Journal of Educational Studies, 67(1), 117–137. https://doi.org/10.1080/00071005.2018.1460870

[6] Voigt, P., & Von dem Bussche, A. (2017). The EU General Data Protection Regulation (GDPR): A practical guide. Springer, Cham. https://doi.org/10.1007/978-3-319-57959-7

[7] Y. Zhang, “Teaching Effectiveness and Student Experience in University Interactive Learning Platforms Based on Mobile Technology”, Int. J. Interact. Mob. Technol., vol. 19, no. 24, pp. pp. 104–119, Dec. 2025.

[8] T. Alasmari, “Artificial Intelligence and M-Learning in Arabic Countries: Innovations, Trends, and Regional Perspectives”, Int. J. Interact. Mob. Technol., vol. 19, no. 05, pp. pp. 170–194, Mar. 2025.

[9] McMahan, B., Moore, E., Ramage, D., Hampson, S., & y Arcas, B. A. (2017). Communication-efficient learning of deep networks from decentralized data. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR 54, 1273–1282.

[10] Li, T., Sahu, A. K., Zaheer, M., Sanjabi, M., Smola, A., & Smith, V. (2020). Federated optimization in heterogeneous networks. Proceedings of Machine Learning and Systems (MLSys), 2, 429–450.

[11] Kairouz, P., McMahan, H. B., Avent, B., Bellet, A., Bennis, M., Bhagoji, A. N., ... & Zhao, S. (2021). Advances and open problems in federated learning. Foundations and Trends® in Machine Learning, 14(1–2), 1–210. https://doi.org/10.1561/2200000083

[12] Lyu, L., Yu, H., & Yang, Q. (2022). Threats to federated learning: A survey. IEEE Transactions on Knowledge and Data Engineering, 36(3), 1951–1972. https://doi.org/10.1109/TKDE.2022.3160699

[13] Diao, E., Ding, J., & Tarokh, V. (2021). HeteroFL: Computation and communication efficient federated learning for heterogeneous clients. Proceedings of the 9th International Conference on Learning Representations (ICLR).

[14] Karimireddy, S. P., Kale, S., Mohri, M., Reddi, S. J., Stich, S. U., & Suresh, A. T. (2020). SCAFFOLD: Stochastic controlled averaging for federated learning. Proceedings of the 37th International Conference on Machine Learning (ICML), PMLR 119, 5132–5143.

[15] Chen, X., Xia, F., & Bekele, A. Z. (2020). Personalized learning path recommendation based on learner knowledge and learning style. IEEE Access, 8, 171206–171222. https://doi.org/10.1109/ACCESS.2020.3024882

[16] Lin, Y., Han, S., Mao, H., Wang, Y., & Dally, W. J. (2018). Deep gradient compression: Reducing the communication bandwidth for distributed training. Proceedings of the 6th International Conference on Learning Representations (ICLR).

[17] Caldas, S., Konecny, J., McMahan, H. B., & Talwalkar, A. (2019). Expanding the reach of federated learning by reducing client resource requirements. arXiv preprint arXiv:1812.07210. https://arxiv.org/abs/1812.07210

[18] Dwork, C., McSherry, F., Nissim, K., & Smith, A. (2006). Calibrating noise to sensitivity in private data analysis. Proceedings of the 3rd Theory of Cryptography Conference (TCC), LNCS 3876, 265–284. https://doi.org/10.1007/11681878_14

[19] Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016). Deep learning with differential privacy. Proceedings of the 23rd ACM Conference on Computer and Communications Security (CCS), 308–318. https://doi.org/10.1145/2976749.2978318

[20] Mironov, I. (2017). Rényi differential privacy. Proceedings of the 30th IEEE Computer Security Foundations Symposium (CSF), 263–275. https://doi.org/10.1109/CSF.2017.11

[21] Geyer, R. C., Klein, T., & Nabi, M. (2017). Differentially private federated learning: A client level perspective. arXiv preprint arXiv:1712.07557. https://arxiv.org/abs/1712.07557

[22] Agarwal, N., Suresh, A. T., Yu, F., Kumar, S., & McMahan, B. (2018). cpSGD: Communication-efficient and differentially-private distributed SGD. Advances in Neural Information Processing Systems (NeurIPS), 31, 7564–7575.

[23] Feldman, V., McMillan, A., & Talwar, K. (2023). Stronger privacy amplification by shuffling for Rényi and approximate differential privacy. Proceedings of the 34th ACM-SIAM Symposium on Discrete Algorithms (SODA), 4966–4981. https://doi.org/10.1137/1.9781611977554.ch182

[24] Corbett, A. T., & Anderson, J. R. (1994). Knowledge tracing: Modeling the acquisition of procedural knowledge. User Modeling and User-Adapted Interaction, 4(4), 253–278. https://doi.org/10.1007/BF01099821

[25] Piech, C., Spencer, J., Huang, J., Ganguli, S., Sahami, M., Guibas, L., & Sohl-Dickstein, J. (2015). Deep knowledge tracing. Advances in Neural Information Processing Systems (NeurIPS), 28, 505–513.

[26] Pandey, S., & Karypis, G. (2019). A self-attentive model for knowledge tracing. Proceedings of the 12th International Conference on Educational Data Mining (EDM), 384–389.

[27] Zou, L., Zhang, H., Liu, Q., & Chen, E. (2020). Instruction-based collaborative exercise recommendation for knowledge training. Proceedings of the 29th ACM International Conference on Information and Knowledge Management (CIKM), 1951–1960. https://doi.org/10.1145/3340531.3411951

[28] Yoo, S., Kim, J. B., & Han, J. (2019). Adaptive mobile learning using context-aware personalization algorithms. IEEE Transactions on Learning Technologies, 12(3), 379–392. https://doi.org/10.1109/TLT.2018.2842076

[29] Nguyen, T., Bhatt, U., Ghassemi, M., & Bhatt, S. (2021). Federated learning for privacy-preserving educational recommendation. Proceedings of the 14th International Conference on Educational Data Mining (EDM), 316–325.

[30] Bettini, C., Riboni, D., & Civitarese, G. (2020). Privacy protection in pervasive systems: State of the art and technical challenges. Pervasive and Mobile Computing, 68, 101212. https://doi.org/10.1016/j.pmcj.2020.101212

[31] Zhao, B., Mopuri, K. R., & Bilen, H. (2020). iDLG: Improved deep leakage from gradients. arXiv preprint arXiv:2001.02610. https://arxiv.org/abs/2001.02610

[32] Smith, V., Chiang, C. K., Sanjabi, M., & Talwalkar, A. S. (2017). Federated multi-task learning. Advances in Neural Information Processing Systems (NeurIPS), 30, 4424–4434.

[33] Fallah, A., Mokhtari, A., & Ozdaglar, A. (2020). Personalized federated learning with theoretical guarantees: A model-agnostic meta-learning approach. Advances in Neural Information Processing Systems (NeurIPS), 33, 3557–3568.

[34] Huang, Y., Chu, L., Zhou, Z., Wang, L., Liu, J., Pei, J., & Zhang, Y. (2021). Personalized cross-silo federated learning on non-IID data. Proceedings of the 35th AAAI Conference on Artificial Intelligence (AAAI), 35(9), 7865–7873. https://doi.org/10.1609/aaai.v35i9.16960