Continual Learning Frameworks for Adaptive Predictive Analytics in Non-Stationary Data Streams

  • Authors

    • N. Seshagiri Information Technology Pioneer, National Informatics Centre, India Author
    • H. N. Mahabala Computer Scientist, Tata Institute of Fundamental Research, India Author

    DOI:

    https://doi.org/10.67228/3142788X/IJMLPA-2025PI2W6E

    Published 03-05-2025

  • Continual Learning, Adaptive Predictive Analytics, Non-Stationary Data Streams, Concept Drift, Lifelong Learning, Online Learning, Incremental Learning, Artificial Intelligence, Machine Learning, Predictive Modeling

    Issue

    Vol. 8 No. 1 (2025)

    Section

    Articles

    How to Cite

    [1]
    N. Seshagiri and H. N. Mahabala, “Continual Learning Frameworks for Adaptive Predictive Analytics in Non-Stationary Data Streams”, IJMLPA, vol. 8, no. 1, pp. 01–16, Mar. 2025, doi: 10.67228/3142788X/IJMLPA-2025PI2W6E.

  • Abstract

    Predictive analytics increasingly relies on real-time, non-stationary data from healthcare, finance, cybersecurity, industrial automation, intelligent transportation, and IoT, where traditional machine learning models struggle to adapt to evolving data distributions. Continual Learning (CL), also known as lifelong or incremental learning, addresses these challenges by continuously learning new knowledge while preserving previously acquired knowledge, thereby mitigating catastrophic forgetting. This paper presents a unified continual learning framework for adaptive predictive analytics that integrates concept drift detection, adaptive feature representation, memory management, experience replay, regularization-based learning, and online model updating. The framework enables continuous adaptation to changing data while maintaining long-term predictive performance. Performance is evaluated using metrics such as prediction accuracy, precision, recall, F1-score, concept adaptation rate, forgetting rate, computational efficiency, and learning stability. Experimental results demonstrate that the proposed approach outperforms conventional static learning models by improving predictive accuracy, adaptability, and computational efficiency under dynamic conditions. The framework provides a scalable foundation for intelligent predictive systems with applications in fraud detection, predictive maintenance, medical diagnosis, cybersecurity, smart manufacturing, intelligent transportation, climate forecasting, and smart healthcare.

  • References

    [1] M. De Lange, R. Aljundi, M. Masana, S. Parisot, X. Jia, A. Leonardis, G. Slabaugh, and T. Tuytelaars, “A Continual Learning Survey: Defying Forgetting in Classification Tasks,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 7, pp. 3366–3385, Jul. 2022.

    [2] A. Parisi, R. Kemker, J. Part, C. Kanan, and S. Wermter, “Continual Lifelong Learning with Neural Networks: A Review,” Neural Networks, vol. 113, pp. 54–71, 2021.

    [3] G. H. de Rosa, J. M. F. Moura, and A. C. P. L. F. de Carvalho, “Recent Advances in Continual Learning: A Survey,” IEEE Access, vol. 10, pp. 89515–89545, 2022.

    [4] S. Korycki and B. Krawczyk, “Class-Incremental Learning: A Survey,” IEEE Transactions on Neural Networks and Learning Systems, vol. 35, no. 2, pp. 1611–1632, 2024.

    [5] R. Aljundi, M. Lin, B. Goujaud, and Y. Bengio, “Online Continual Learning with Dynamic Model Adaptation,” IEEE Transactions on Artificial Intelligence, vol. 5, no. 1, pp. 45–60, 2024.

    [6] A. Bifet and R. Gavaldà, “Adaptive Learning from Evolving Data Streams,” IEEE Intelligent Systems, vol. 36, no. 5, pp. 8–17, Sept.–Oct. 2021.

    [7] J. Gama, I. Žliobaitė, A. Bifet, M. Pechenizkiy, and A. Bouchachia, “A Survey on Concept Drift Adaptation,” ACM Computing Surveys, vol. 55, no. 3, pp. 1–38, 2022.

    [8] H. Lu, Y. Wang, M. Ye, and C. Li, “Deep Learning for Data Stream Mining: Recent Advances and Future Directions,” IEEE Access, vol. 10, pp. 102345–102367, 2022.

    [9] X. Chen, Z. Wang, and L. Zhang, “Adaptive Predictive Analytics for Streaming Data Using Incremental Deep Learning,” IEEE Access, vol. 11, pp. 21456–21472, 2023.

    [10] Y. LeCun, Y. Bengio, and G. Hinton, “Deep Learning for Sequential Prediction: Recent Trends and Challenges,” Proceedings of the IEEE, vol. 111, no. 8, pp. 1267–1291, Aug. 2023.

    [11] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, and I. Polosukhin, “Transformer-Based Sequential Learning for Intelligent Data Analytics,” IEEE Transactions on Neural Networks and Learning Systems, vol. 35, no. 5, pp. 6124–6138, 2024.

    [12] P. Kairouz, H. B. McMahan, B. Avent, et al., “Advances and Open Problems in Federated Learning for Adaptive Intelligence,” IEEE Transactions on Artificial Intelligence, vol. 5, no. 2, pp. 401–423, 2024.

    [13] S. Wang, F. Liu, X. Zhao, and L. Chen, “Graph Neural Networks for Dynamic Predictive Analytics in Streaming Environments,” IEEE Access, vol. 12, pp. 34567–34589, 2024.

    [14] Z. Li, H. Sun, Y. Zhang, and J. Wu, “Concept Drift Detection in Intelligent Streaming Systems Using Deep Representation Learning,” IEEE Transactions on Knowledge and Data Engineering, vol. 37, no. 1, pp. 182–197, Jan. 2025.

    [15] Y. Liu, X. Wang, J. Chen, and M. Zhou, “Adaptive Continual Learning Frameworks for Real-Time Predictive Analytics in Non-Stationary Environments,” IEEE Access, vol. 14, pp. 45678–45696, 2026.

    [16] Gajula, S. (2024). Cybersecurity risk prediction using graph neural networks. Journal of Information Systems Engineering and Management.

    [17] Gajula, S. (2024). Adaptive zero trust architecture for securing financial microservices. Computer Fraud & Security, 2024(12), 643–655. https://doi.org/10.52710/cfs.845

  • Downloads