What question did this study set out to answer?

The aim is to review the evolution of cyber threat detection methods using federated machine learning while addressing privacy concerns and scalability.

May 8, 2026Open Access

A Comprehensive Literature Review on Federated Machine Learning for Privacy-Preserving Cyber Threat Detection in Distributed Network Environments

Key Points

The aim is to review the evolution of cyber threat detection methods using federated machine learning while addressing privacy concerns and scalability.
Comprehensive literature review on federated machine learning techniques related to cyber threat detection.
Analysis of the evolution from rule-based systems to advanced machine learning and deep learning techniques.
Examination of challenges in federated learning, including data variability, communication overhead, and security threats.
Identified key advancements in privacy protection through federated learning approaches.
Highlighted persistent challenges such as the need for improved communication and security systems.
Pointed out existing research gaps in achieving practical scalable solutions for real-world applications.

Abstract

Cloud computing and IoT devices are actually growing very fast, and this definitely makes cyber attacks more complex and common. Traditional systems for catching cyber attacks actually have problems with new threats and keeping data safe. These old methods definitely cannot handle big amounts of data spread across many places. This paper gives a complete study review regarding federated machine learning for keeping privacy safe in cyber threat detection as per distributed network systems. The study examines how cyber threat detection methods have evolved from basic rule-based systems to advanced machine learning approaches. It further analyzes how the field itself has progressed from simple anomaly detection to complex deep learning techniques. These methods surely make detection more accurate, but they depend too much on processing data in one central place. Moreover, this creates problems with privacy protection and handling large amounts of data. Federated learning actually solves these problems by letting different computers work together to train models using their own data. The computers definitely learn together but never actually share their raw information with each other. As per this method, data privacy gets better regarding protection, and the system becomes more scalable and strong. The review actually looks at important methods in federated learning like secure combining, privacy protection, and coding systems that definitely make the system more safe. Also, this study actually looks at the main problems in federated learning like different types of data, too much communication, and attacks from bad actors. These challenges definitely make the system harder to work with. Basically, the study shows the same research gaps and says we need good communication methods, strong security systems, and scalable designs for real-world use. We are seeing that federated learning can only change cybersecurity by helping different systems work together to find threats while keeping data safe and private.

A Comprehensive Literature Review on Federated Machine Learning for Privacy-Preserving Cyber Threat Detection in Distributed Network Environments

Key Points

Abstract

Cite This Study