Federated Learning for Privacy-Preserving Fraud Detection in Digital Banking: Balancing Algorithmic Performance, Privacy, and Regulatory Compliance

@article{1709491,
      author = {Michael Friday Umakor, Ikechukwu Iheanyi, Ugochukwu Daniel Ofurum, Ugochukwu Henry Ben Ibecheozor, Elizabeth Ayodeji Adeyefa},
      title = {Federated Learning for Privacy-Preserving Fraud Detection in Digital Banking: Balancing Algorithmic Performance, Privacy, and Regulatory Compliance},
      journal = {Iconic Research And Engineering Journals},
      year = {2025},
      volume = {9},
      number = {1},
      pages = {215-231},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1709491.pdf},
      abstract = {The rapid growth of digital banking has heightened concerns over cybersecurity, privacy, and regulatory compliance, particularly in the detection and prevention of financial fraud. Traditional centralized machine learning approaches to fraud detection are limited by data privacy regulations and the increasing complexity of cyber threats. Federated Learning (FL), a decentralized machine learning technique, offers a promising alternative by enabling multiple institutions to collaboratively train models without sharing raw data. This study critically evaluates the application of FL in privacy-preserving fraud detection within the banking sector, focusing on algorithmic performance, privacy implications, and regulatory compliance. The paper reviews existing literature, assesses technical challenges such as data heterogeneity and communication overhead, and presents case studies of FL implementation in real-world banking contexts. The findings reveal that FL significantly enhances privacy and regulatory alignment while maintaining competitive fraud detection performance. The study concludes by offering strategic recommendations for digital banks and regulatory bodies and identifies future research directions that emphasize adaptive learning algorithms, robust evaluation frameworks, and long-term federated infrastructure in financial systems.},
      keywords = {Federated Learning; Fraud Detection; Digital Banking; Privacy; Regulatory Compliance; Machine Learning},
      month = {July},
  }