FEDERATED LEARNING-BASED PRIVACY PRESERVATION FOR INTELLIGENT DATA ANALYTICS
TL;DR
The authors discussed the potential of federated learning in intelligent data analytics with privacy, and reported the potential advantages of federated learning in the context of secure and intelligent data analysis, while preserving data privacy.
Abstract
Federated Learning (FL) is a new machine learning paradigm that allows different users and organizations to jointly learn an intelligent model without sharing local private data. Federated learning retains data locally on devices and sends only model updates to a central server to prevent the owners from losing their data. It can dramatically curtail confidential risks to sensitive information across a variety of sectors, including healthcare, finance, smart cities, and Internet of Things (IoT) use. In this paper, the authors discuss the potential of federated learning in intelligent data analytics with privacy. Briefly introduces the key concepts, current research, privacy concerns, communications issues, and security solutions for a federated environment. The results reported the potential advantages of federated learning in the context of secure and intelligent data analysis, while preserving data privacy. Keywords: Federated Learning, Data Privacy, Intelligent Data Analytics, Machine Learning, Data Security, Privacy Preservation
