Loading...
Jayakeerthi M. is an Assistant Professor in the Department of IoT and Artificial Intelligence & Machine Learning at Nehru Arts and Science College, India. Her research interests include IoT, Artificial Intelligence, and smart computing technologies.
The rising quantity of digital financial transactions in banking, e-commerce, mobile payment systems, cloud computing, and Internet of Things (IoT) environments has heightened the demand for sophisticated fraud detection mechanisms. Traditional machine learning algorithms necessitate the centralised aggregation of sensitive consumer and transaction information, raising issues related to data privacy, regulatory adherence, and cybersecurity. Federated Learning (FL) has arisen as a privacy-centric distributed machine learning framework that allows various organisations or devices to jointly develop fraud detection models while safeguarding raw data. Rather than sharing sensitive data, just model parameters or gradients are sent, thus mitigating privacy concerns while preserving analytical efficacy. When integrated with Artificial Intelligence (AI), Differential Privacy, Secure Multi-Party Computation (SMPC), Homomorphic Encryption, and Blockchain, Federated Learning establishes a secure and scalable infrastructure for cooperative fraud analysis. This chapter elucidates the ideas, framework, methodologies, applications, obstacles, and prospective advancements of Federated Learning for safeguarding privacy in fraud analysis within contemporary financial systems.