Federated Learning for Cybersecurity: Concepts, Challenges and Future Directions

Mamoun Alazab, Swarna Priya R M, Parimala M, Praveen Reddy, Thippa Reddy Gadekallu, Quoc Viet Pham

    Research output: Contribution to journalArticlepeer-review

    Abstract

    Federated Learning (FL) is a recent development in Artificial Intelligence (AI) which is based on the concept of decentralized data. Cyber-attacks are frequently happening in various applications that are deployed in real-time scenarios due to which most of the industrialists hesitate to move forward in adopting the technology of Internet of Everything (IoE). This paper aims to provide an extensive study on how FL could be utilized for providing better cybersecurity and prevent various cyber-attacks in real-time. We present an extensive survey of the various FL models currently developed by researchers for providing authentication, privacy, trust management and attack detection. We also discuss few real-time use cases that are deployed recently and how FL is adopted in them for providing privacy of data. Based on the study, we conclude the paper with some prominent challenges and future directions on which the researchers can focus for adopting FL in real-time scenarios.

    Original languageEnglish
    Number of pages6
    JournalIEEE Transactions on Industrial Informatics
    DOIs
    Publication statusE-pub ahead of print - Oct 2021

    Fingerprint

    Dive into the research topics of 'Federated Learning for Cybersecurity: Concepts, Challenges and Future Directions'. Together they form a unique fingerprint.

    Cite this