Graph-Based Federated Learning in Distributed Cloud Ecosystems for Scalable Big Data Analytics

Authors

  • Sai Kiran Reddy Malikireddy Department of Engineering, University of South Florida Author

Keywords:

Graph-Based Federated Learning, Distributed Cloud Ecosystems, Big Data Analytics, Graph Neural Networks (GNNs), Distributed Graph Processing, Scalable Machine Learning.

Abstract

The exponential growth of data in distributed cloud environments has necessitated scalable and secure solutions for analytics. Federated learning, a privacy-preserving machine learning paradigm, provides a robust approach by enabling decentralized model training without direct data sharing. Integrating graph-based techniques into federated learning enhances its capability to analyze structured data, such as social networks, sensor networks, and interconnected IoT systems. This paper proposes a graph-based federated learning framework tailored for distributed cloud ecosystems, enabling efficient processing and analysis of large-scale big data. By leveraging graph neural networks (GNNs) and distributed graph processing, the framework achieves scalability, improves model accuracy, and maintains data confidentiality. Experimental results on real-world datasets demonstrate significant performance gains in handling heterogeneous and dynamic data across geographically dispersed nodes. This work highlights the potential of combining graph theory with federated learning to advance big data analytics in distributed cloud environments.

Downloads

Download data is not yet available.

Downloads

Published

2024-11-21

Similar Articles

1-10 of 231

You may also start an advanced similarity search for this article.