Privacy-Preserving Data Engineering in Financial Services: Differential Privacy, Federated Learning, and Secure Computation Models
DOI:
https://doi.org/10.63282/3050-9262.IJAIDSML-V2I4P110Keywords:
Privacy-Preserving Data Engineering, Differential Privacy, Federated Learning, Secure Multi-Party Computation, Data Protection, Data Masking, TokenizationAbstract
Financial institutions process highly sensitive personal and transactional data, making privacy protection a critical requirement for modern data platforms. With increasing regulatory scrutiny under frameworks such as GDPR, CCPA, and global financial compliance standards, organizations must adopt advanced techniques that protect customer data while enabling meaningful analytics and machine learning. Traditional approaches such as data masking and tokenization provide partial protection but often limit the usability of datasets for advanced analytics. This paper explores privacy-preserving data engineering techniques that enable secure data processing without exposing sensitive information. The study examines differential privacy mechanisms, federated learning architectures, and secure multi-party computation models as viable approaches for privacy-aware financial analytics. The paper proposes a layered privacy architecture that integrates these techniques within modern financial data pipelines to ensure secure data processing while preserving analytical value. Through architectural analysis and practical implementation considerations, this research demonstrates how financial institutions can balance privacy protection, regulatory compliance, and data-driven innovation
References
[1] Art. 3 GDPRTerritorial scope, Intersoft Consulting, online. https://gdpr-info.eu/art-3-gdpr/
[2] GDPRFines / Penalties, Intersoft Consulting, online. https://gdpr-info.eu/issues/fines-penalties/
[3] California Consumer Privacy Act (CCPA), State of California Department of Justice. online. https://www.oag.ca.gov/privacy/ccpa
[4] Gramm-Leach-Bliley Act, Federal Trade Commission. online. https://www.ftc.gov/business-guidance/privacy-security/gramm-leach-bliley-act
[5] Privacy Rule Handbook, FDIC. online. https://www.fdic.gov/bank-examinations/privacy-rule-handbook
[6] CFPB Finalizes Personal Financial Data Rights Rule to Boost Competition, Protect Privacy, and Give Families More Choice in Financial Services, CFPB. online. https://www.consumerfinance.gov/about-us/newsroom/cfpb-finalizes-personal-financial-data-rights-rule-to-boost-competition-protect-privacy-and-give-families-more-choice-in-financial-services/
[7] PCI Data Security Standard (PCI DSS), Security Standard Council, online. https://listings.pcisecuritystandards.org/documents/Tokenization_Guidelines_Info_Supplement.pdf
[8] What is data masking? Types, techniques and best practice, ISO. online. https://www.iso.org/information-security/data-masking
[9] special Publication 800-12: An Introduction to Computer Security: The NIST Handbook, online. https://csrc.nist.rip/publications/nistpubs/800-12/800-12-html/chapter19.html
[10] The Algorithmic Foundations of Differential Privacy, Foundations and Trends. online. https://www.cis.upenn.edu/~aaroth/Papers/privacybook.pdf
[11] Collaboration Space, Privacy Engineering Program, NIST. Online. https://www.nist.gov/itl/applied-cybersecurity/privacy-engineering/collaboration-space
[12] Learning with Privacy at Scale, Machine Learning Research, 2017. online. https://machinelearning.apple.com/research/learning-with-privacy-at-scale
[13] McMahan, B., Moore, E., Ramage, D., Hampson, S., & y Arcas, B. A. (2017, April). Communication-efficient learning of deep networks from decentralized data. In Artificial intelligence and statistics (pp. 1273-1282). Pmlr.
[14] Kairouz, P., & McMahan, H. B. (2021). Advances and open problems in federated learning. Foundations and trends in machine learning, 14(1-2), 1-210.
[15] Federated Learning: Collaborative Machine Learning without Centralized Training Data, Google Research. online. https://research.google/blog/federated-learning-collaborative-machine-learning-without-centralized-training-data/
[16] The Foundations of Cryptography - Volume 2, Cambridge University Press, online. https://www.wisdom.weizmann.ac.il/~oded/foc-vol2.html
[17] Evans, D., Kolesnikov, V., & Rosulek, M. (2018). A pragmatic introduction to secure multi-party computation. Foundations and Trends® in Privacy and Security, 2(2-3), 70-246.
[18] Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019). Federated machine learning: Concept and applications. ACM Transactions on Intelligent Systems and Technology, 10(2), 1–19.
[19] Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2020). Federated learning: Challenges, methods, and future directions. IEEE Signal Processing Magazine, 37(3), 50–60.
[20] Dwork, C., & Roth, A. (2014). The algorithmic foundations of differential privacy. Foundations and trends® in theoretical computer science, 9(3-4), 211-487.
[21] Bonawitz, K., Ivanov, V., Kreuter, B., Marcedone, A., McMahan, H. B., Patel, S., ... & Seth, K. (2016). Practical secure aggregation for federated learning on user-held data. arXiv preprint arXiv:1611.04482.
