Sustainable (green) cloud infrastructure for ML‑enhanced fraud detection

  • Authors

    • Rajesh Kumar Sharma Professor, University of Delhi, India Author
    • Priya Natarajan Associate Professor, University of Madras, India Author

    DOI:

    https://doi.org/10.67228/30713315/IJAIDT-2022PI3J7F

    Published 05-03-2022

  • Sustainable Computing, Green Cloud Infrastructure, Machine Learning, Fraud Detection, Cloud Computing, Carbon-Aware Computing, Energy Efficiency, Responsible AI, Green AI, Eco-Friendly ML Systems

    Issue

    Vol. 5 No. 1 (2022)

    Section

    Articles

    How to Cite

    [1]
    R. K. Sharma and P. Natarajan, “Sustainable (green) cloud infrastructure for ML‑enhanced fraud detection”, IJAIDT, vol. 5, no. 1, pp. 01–15, May 2022, doi: 10.67228/30713315/IJAIDT-2022PI3J7F.

  • Abstract

    As machine learning becomes central to modern fraud detection systems, the computational and environmental demands of these solutions have surged. While cloud infrastructure enables scalable deployment of ML models, it also contributes significantly to global energy consumption and carbon emissions. This paper explores the integration of sustainable (green) cloud computing practices into ML-enhanced fraud detection systems. We analyze the environmental impact of cloud-based ML workloads and identify principles and strategies for building energy-efficient, carbon-aware infrastructure. By proposing a sustainable architecture for fraud detection, we aim to balance high detection accuracy with minimized ecological footprint. The paper concludes with practical guidelines, challenges, and future research directions for achieving greener fraud detection systems

  • References

    [1] Strubell, E., Ganesh, A., & McCallum, A. (2019). Energy and policy considerations for deep learning in NLP. ACL.

    [2] Schwartz, R., Dodge, J., Smith, N. A., & Etzioni, O. (2020). Green AI. Communications of the ACM, 63(12), 54–63.

    [3] Peng, C., Li, B., & Zhang, Q. (2022). Carbon-aware computing: Techniques and challenges. ACM Computing Surveys, 55(1), 1–36.

    [4] Vögler, M., Inzinger, C., & Dustdar, S. (2021). Energy-efficient cloud computing. IEEE Cloud Computing, 8(2), 18–26.

    [5] IEEE. (2022). Green AI and cloud systems: A call for sustainable ML. IEEE Access.

    [6] ISO. (2015). ISO 14001:2015 - Environmental management systems — Requirements with guidance for use.

    [7] Chen, J., Liu, H., & Chen, X. (2021). Edge AI for real-time fraud detection. Journal of Parallel and Distributed Systems, 152, 35–44.

    [8] Zhou, Y., & Yu, Y. (2020). Federated learning for privacy-preserving fraud detection in mobile banking. IEEE Transactions on Services Computing.

    [9] Hölzle, U., & Barroso, L. A. (2013). The datacenter as a computer: An introduction to the design of warehouse-scale machines. Synthesis Lectures on Computer Architecture.

    [10] Patterson, D., Gonzalez, J., & Le, Q. V. (2021). Carbon emissions and large neural networks. NeurIPS Workshop on Tackling Climate Change.

  • Downloads