Federated Learning-Based Personalized Recommendation Systems: An Overview on Security and Privacy Challenges
Abstract
The recent advancement in next-generation Consumer Electronics (CE) has created the problems of information overload and information loss. The significance of Personalized Recommendation Systems (PRS) to efficiently and effectively extract useful user information is seen as an ideal solution to provide users with personalized content and services and therefore is used in different application domains including healthcare, e-commerce, social media, etc. Security and privacy are the two major challenges of the existing PRS for next-gen CE data. Federated learning (FL) has the potential to elevate the aforementioned challenges by sharing local recommender parameters while keeping all the training data on the device and therefore is seen as a promising technique to enhance security and privacy in PRS for the next-gen CE data. In this survey, we have first discussed the enhancement of the existing CE technologies, a holistic review of security and privacy challenges in current PRS, and the advantage of FL-based PRS for next-gen CE. Finally, we list a few open issues and challenges that can guide researchers and practitioners to further drive research in this promising area.
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.