Privacy-Preserving Personalization Using Federated Learning in AEM
DOI:
https://doi.org/10.63282/3050-9416.IJAIBDCMS-V4I4P119Keywords:
Privacy-Preserving Personalization, Federated Learning, AEM, Adobe Experience Manager, Edge AI, User Data Privacy, GDPR Compliance, On-Device Learning, Machine Learning, Digital Experience PlatformsAbstract
Adobe Experience Manager (AEM) relies heavily on personalized digital experiences, as they are a key enabler of brands distributing the right, timely, and user-friendly content that matches each individual customer's needs. On the other hand, reaching that high level of personalization frequently entails the gathering and processing of sensitive data located on the client side, thereby posing privacy problems of user tracking, data centralization, and conformity with rules such as GDPR and CCPA. Consequently, Federated Learning (FL) stands out as an excellent alternative in that it permits the training of machine-learning models to be done on the user's device without having to send the raw personal data to a central server. This paper devises a privacy-respecting personalization mechanism that uses FL to communicate with AEM's personalization engine, consequently allowing a company to obtain behavioral insights without violating the privacy of the users. Herein, the AEM-managed digital touchpoints receive the deployment of lightweight FL models that learn from on-device interactions, and at regular intervals, they send only the encrypted model updates to the server, thus ensuring privacy preservation. From there, AEM aggregates these updates to fine-tune global personalization models that are sent back to the devices for further training. As such, the method retains the sumptuousness of AEM personalization features like content targeting, segmentation, and predictive recommendations while simultaneously achieving a significant cut in privacy risk. The suggested remedy achieves refined model metrics, augmented user engagement, and shorter personalization response time due to execution at the local level, all happening concurrently with the maintenance of data secrecy. The potential of such a solution includes the optimization of e-commerce, the scoring of content relevance, real-time audience segmentation, and adaptive customer journeys across websites and mobile apps.
References
1. Ali, Mansoor, et al. "Federated learning for privacy preservation in smart healthcare systems: A comprehensive survey." IEEE journal of biomedical and health informatics 27.2 (2022): 778-789.
2. Wang, Yi, Ning Gao, and Gabriela Hug. "Personalized federated learning for individual consumer load forecasting." CSEE Journal of Power and Energy Systems 9.1 (2022): 326-330.
3. Liu, Bingyan, Yao Guo, and Xiangqun Chen. "PFA: Privacy-preserving federated adaptation for effective model personalization." Proceedings of the Web Conference 2021. 2021.
4. Loftus, Tyler J., et al. "Federated learning for preserving data privacy in collaborative healthcare research." Digital Health 8 (2022): 20552076221134455.
5. Ruzafa-Alcázar, Pedro, et al. "Intrusion detection based on privacy-preserving federated learning for the industrial IoT." IEEE Transactions on Industrial Informatics 19.2 (2021): 1145-1154.
6. Rischke, Roman, et al. "Federated learning in dentistry: chances and challenges." Journal of dental research 101.11 (2022): 1269-1273.
7. Mansour, Yishay, et al. "Three approaches for personalization with applications to federated learning." arXiv preprint arXiv:2002.10619 (2020).
8. Li, Zijian, et al. "Hpfl-cn: Communication-efficient hierarchical personalized federated edge learning via complex network feature clustering." 2022 19th Annual IEEE International Conference on Sensing, Communication, and Networking (SECON). IEEE, 2022.
9. Parakala, Adityamallikarjunkumar, and Jyothirmay Swain. "AI‑Powered Intelligent Automation Emerges." International Journal of Artificial Intelligence, Data Science, and Machine Learning 3.4 (2022): 96-106.
10. Bharati, Subrato, et al. "Federated learning: Applications, challenges and future directions." International Journal of Hybrid Intelligent Systems 18.1-2 (2022): 19-35.
11. Demertzis, Konstantinos, et al. "An explainable semi-personalized federated learning model." Integrated Computer-Aided Engineering 29.4 (2022): 335-350.
12. Sun, Peng, et al. "A profit-maximizing model marketplace with differentially private federated learning." IEEE INFOCOM 2022-IEEE Conference on Computer Communications. IEEE, 2022.
13. Chen, Tianyu, et al. "The-x: Privacy-preserving transformer inference with homomorphic encryption." arXiv preprint arXiv:2206.00216 (2022).
14. Qi, Jiahao, et al. "High-quality model aggregation for blockchain-based federated learning via reputation-motivated task participation." IEEE Internet of Things Journal 9.19 (2022): 18378-18391.
15. Saha, Sudipan, and Tahir Ahmad. "Federated transfer learning: Concept and applications." Intelligenza Artificiale 15.1 (2021): 35-44.
16. Yu, Shuai, et al. "When deep reinforcement learning meets federated learning: Intelligent multitimescale resource management for multiaccess edge computing in 5G ultradense network." IEEE Internet of Things Journal 8.4 (2020): 2238-2251.
