A Differential Privacy-Based Learning Path Optimization Algorithm for Distributed-Network Industry Operation and Maintenance Skills

2026

Volume 35

2026-09-16

TaiHui Mao^1,2, Junlin Wu³, and Yining Kang⁴

¹Beijing Lightyear Infinite Technology Co., Ltd (Turing robot), Beijing, 100000, China

²Jiangxi Modern Polytechnic College, Nanchang Jiangxi, 330000, China

³Hangzhou Banyun Technology Co., Ltd, Hangzhou Zhejiang, 310051, China

⁴Smart Whale Tech (HK) Limited, Hong Kong, 999077, China

Received: July 20, 2026
Accepted: August 16, 2026
Publication Date: September 16, 2026

Distributed privacy-preserving learning path optimization system architecture

Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: BibTeX | http://dx.doi.org/10.6180/jase.202612_35.040

Download PDF

In distributed network operation and maintenance (O&M) systems, skill-training behavior data are scattered across enterprise sites, training platforms, edge servers, and field terminals, so centralized learning-path optimization risks exposing sensitive information such as engineers’ response feedback, weak skill points, and training stages. This study develops DP-DLPO, a differential privacy-based distributed learning path optimization algorithm with five modules: local training-state encoding, learning path graph modeling, differential privacy perturbation, adaptive privacy budget allocation, and quality-aware federated aggregation. Each edge node uploads only model updates after gradient pruning and noise perturbation, and the cloud server aggregates updates by sample size, update quality, and noise intensity to generate a global path optimization strategy. Simulation results show DP-DLPO achieves an F1-score of 0.854, NDCG@10 of 0.817, and path completion rate of 0.843, outperforming DP-FedAvg and FedRec-DP. With privacy budget ε=2, the model maintains strong performance while keeping attack success at 0.094, and as nodes increase from 10 to 200, F1 decreases only from 0.858 to 0.833—showing good scalability and path interpretability for privacy-preserving skill-path optimization in industrial network O&M systems.

Keywords: differential privacy; learning path optimization; distributed networks; federated learning; industrial operation and maintenance; skill-training data mining

[1] P. Kairouz and H. B. McMahan, (2021) “Advances and Open Problems in Federated Learning” Foundations and Trends in Machine Learning 14(1-2): 1-210. DOI: 10.1561/2200000083.
[2] Q. Li, Z. Wen, Z. Wu, S. Hu, N. Wang, Y. Li, X. Liu, and B. He, (2023) “A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection” IEEE Transactions on Knowledge and Data Engineering 35(4): 3347-3366. DOI: 10.1109/TKDE.2021.3124599.
[3] B. Liu, N. Lv, Y. Guo, and Y. Li, (2024) “Recent Advances on Federated Learning: A Systematic Survey” Neurocomputing 597: 128019. DOI: 10.1016/j.neucom.2024.128019.
[4] V. Mothukuri, R. M. Parizi, S. Pouriyeh, Y. Huang, A. Dehghantanha, and G. Srivastava, (2021) “A Survey on Security and Privacy of Federated Learning” Future Generation Computer Systems 115: 619-640. DOI: 10.1016/j.future.2020.10.007.
[5] A. El Ouadrhiri and A. Abdelhadi, (2022) “Differential Privacy for Deep and Federated Learning: A Survey” IEEE Access 10: 22359-22380. DOI: 10.1109/ACCESS.2022.3151670.
[6] X. Ma, J. Zhu, Z. Lin, S. Chen, and Y. Qin, (2022) “A State-of-the-Art Survey on Solving Non-IID Data in Federated Learning” Future Generation Computer Systems 135: 244-258. DOI: 10.1016/j.future.2022.05.003.
[7] L. Lyu, H. Yu, X. Ma, C. Chen, L. Sun, J. Zhao, Q. Yang, and P. S. Yu, (2024) “Privacy and Robustness in Federated Learning: Attacks and Defenses” IEEE Transactions on Neural Networks and Learning Systems 35(7): 8726-8746. DOI: 10.1109/TNNLS.2022.3216981.
[8] C. Chronis, I. Varlamis, Y. Himeur, A. N. Sayed, T. M. Al-Hasan, A. Nhlabatsi, F. Bensaali, and G. Dimitrakopoulos, (2024) “A Survey on the Use of Federated Learning in Privacy-Preserving Recommender Systems” IEEE Open Journal of the Computer Society 5: 227-247. DOI: 10.1109/OJCS.2024.3396344.
[9] Z. Xu, C. Chu, and S. Song, (2024) “An Effective Federated Recommendation Framework with Differential Privacy” Electronics 13(8): 1589. DOI: 10.3390/electronics13081589.
[10] Q. Guo, F. Zhuang, C. Qin, H. Zhu, X. Xie, H. Xiong, and Q. He, (2022) “A Survey on Knowledge Graph-Based Recommender Systems” IEEE Transactions on Knowledge and Data Engineering 34(8): 3549-3568. DOI: 10.1109/TKDE.2020.3028705.
[11] C. Gao, Y. Zheng, N. Li, Y. Li, Y. Qin, J. Piao, Y. Quan, J. Chang, D. Jin, X. He, and Y. Li, (2023) “A Survey of Graph Neural Networks for Recommender Systems: Challenges, Methods, and Directions” ACM Transactions on Recommender Systems 1(1): 1-51. DOI: 10.1145/3568022.
[12] Y. Li, D. Chen, and Z. Zhan, (2022) “Research on Personalized Recommendation of MOOC Resources Based on Ontology” Interactive Technology and Smart Education 19(4): 422-440. DOI: 10.1108/ITSE-10-2021-0190.
[13] J. Wang, H. Xie, F. L. Wang, L.-K. Lee, and O. T. S. Au, (2021) “Top-N Personalized Recommendation with Graph Neural Networks in MOOCs” Computers and Education: Artificial Intelligence 2: 100010. DOI: 10.1016/j.caeai.2021.100010.
[14] X. Zhang, S. Liu, and H. Wang, (2023) “Personalized Learning Path Recommendation for E-Learning Based on Knowledge Graph and Graph Convolutional Network” International Journal of Software Engineering and Knowledge Engineering 33(1): 109-131. DOI: 10.1142/S0218194022500681.
[15] L.-Y. Zhou and Y.-Y. Wang, (2025) “Simulation of Personalized English Learning Path Recommendation System Based on Knowledge Graph and Deep Reinforcement Learning” Scientific Reports 15: 34554. DOI: 10.1038/s41598-025-17918-x.