A Review on Advancing Cybersecurity Frameworks by Integrating Machine Intelligence with Federated Learning
Vijayta Patil, Jhankar Moolchandani, Ashutosh Kumar Dubey · 2024
In today's world, the growth of connected networks has also created a significant risk due to increasing risks of cyber-attacks. In this paper, various hybrid cybersecurity frameworks have been reviewed and analyzed that utilize machine learning approaches in combination with federated learning. These frameworks are assessed in terms of their security against complex cyber threats. In particular, machine learning focuses on improving threats detection by searching for patterns and other oddities in complex datasets, while the federated learning decentralizes the topology of the model training. This decentralization helps to enhance strict data compliance requirements and reduce data exposure risks. When synergized in the dual approach, the security system is strengthened as well as the enforcement of data privacy and other compliance where available. Empirical results have proved that this methodological approach provides better outcomes especially aimed at detection among other benchmarks. The improvement due to the hybridization indicates a scalable and robust solution that can adapt the evolving cyber threats landscape. The hybrid combination may establish new cybersecurity standards and open the door to more secure digital interactions across various sectors.