A Comprehensive Research of Data Privacy Based on Federated Learning

Junxiang Zhang

2024

Abstract

In recent years, Federated Learning (FL) has gained significant attention as a crucial technology for addressing the issue of data silos. Despite possessing certain privacy-preserving capabilities, FL still carries the risk of privacy leakage, particularly in fields such as healthcare and finance, where the demand for user privacy protection is increasingly urgent. This review first introduces the fundamental principles and classifications of FL, with a focus on discussing its advantages in data privacy protection. Subsequently, it reviews the background of current data privacy challenges, encompassing various privacy attack methods that highlight the deficiencies of FL in privacy protection. Following this, various privacy protection methods are thoroughly discussed, analyzing the strengths of different methods in safeguarding data privacy. A comparative analysis of specific privacy protection algorithms is then conducted, providing a detailed examination of the advantages, disadvantages, protection strategies, and targeted subjects of each algorithm. By systematically summarizing existing research, this paper offers a comprehensive understanding of the application of FL in the field of data privacy, providing valuable insights for both the academic and industrial sectors. Furthermore, it serves as a useful guide for future research and applications in this domain.

Paper Citation

in Harvard Style

Zhang J. (2024). A Comprehensive Research of Data Privacy Based on Federated Learning. In Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE; ISBN 978-989-758-690-3, SciTePress, pages 531-536. DOI: 10.5220/0012832500004547

in Bibtex Style

@conference{icdse24,
author={Junxiang Zhang},
title={A Comprehensive Research of Data Privacy Based on Federated Learning},
booktitle={Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE},
year={2024},
pages={531-536},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012832500004547},
isbn={978-989-758-690-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Data Science and Engineering - Volume 1: ICDSE
TI - A Comprehensive Research of Data Privacy Based on Federated Learning
SN - 978-989-758-690-3
AU - Zhang J.
PY - 2024
SP - 531
EP - 536
DO - 10.5220/0012832500004547
PB - SciTePress