Evolutionary Adaptation of DLP Policies under Concept Drift in Streaming Data
DOI:
https://doi.org/10.32515/2664-262X.2025.12(43).2.9-19Keywords:
genetic algorithms, Data Loss Prevention, anomaly detection, concept drift, cloud securityAbstract
In modern streaming DLP systems deployed across cloud and hybrid environments, fixed policies degrade rapidly due to concept drift. Operators must simultaneously control the risk-weighted miss cost, limit the false- alarm burden, meet latency SLOs, and keep alert streams stable under tight memory and compute budgets. These competing objectives are not adequately balanced by traditional detectors or manual policy tuning.
We present an online evolutionary controller that casts policy adaptation as constrained multi-objective optimization. The method uses a chromosome encoding with drift-aware exploration–exploitation switching, an archive of vetted policies for warm starts, a compact active mixture, and guarded rollbacks for operational safety. On six streams (synthetic and real), the controller keeps the integrated cost within 0–3.5% of the best baseline (mean absolute gap ≈1.6%), sustains p95 latency below 100 ms, and reduces alert-rate volatility by 50–63% while maintaining comparable or lower false-alarm rates.
Two practical sensitivities emerge: the drift-gate threshold governing the exploration/exploitation balance, and short-lived compute bursts immediately after detected changes. Warm starts, a compact mixture, and mutation-budget guards mitigate these effects without sacrificing responsiveness.
References
Список літератури
1. Hinder F., Vaquet V., Hammer B. One or two things we know about concept drift – A survey on monitoring in evolving environments. Part A: Detecting concept drift. Frontiers in Artificial Intelligence. 2024. Vol. 7. Art. 1330257. DOI: https://doi.org/10.3389/frai.2024.1330257.
2. Li W., Yang X., Liu W., Xia Y., Bian J. DDG-DA: Data Distribution Generation for Predictable Concept Drift Adaptation. Proceedings of the AAAI Conference on Artificial Intelligence. 2022. Vol. 36, № 4. С. 4092–4100. DOI: https://doi.org/10.1609/aaai.v36i4.20327.
3. Baier L., Schlör T., Schöffer J., Kühl N. Detecting Concept Drift with Neural Network Model Uncertainty. arXiv preprint. 2021. DOI: https://doi.org/10.48550/arXiv.2107.01873.
4. Alneyadi S., Sithirasenan E., Muthukkumarasamy V. A Survey on Data Leakage Prevention Systems. Journal of Network and Computer Applications. 2016. Vol. 62. С. 137–152. DOI: https://doi.org/10.1016/j.jnca.2016.01.008.
5. de Barros R. S. M., Santos S. G. T. C. A Large-Scale Comparison of Concept Drift Detectors. Information Sciences. 2018. Vol. 451–452. С. 348–370. DOI: https://doi.org/10.1016/j.ins.2018.04.014.
6. Losing V., Hammer B., Wersing H. Drift Datasets. GitHub repository. 2025. URL: https://github.com/vlosing/driftDatasets (дата звернення: 05.11.2025).
7. Gomes H. M., Bifet A., Read J., Barddal J. P., Enembreck F., Pfahringer B., Holmes G., Abdessalem T. Adaptive Random Forests for Evolving Data Stream Classification. Machine Learning. 2017. Vol. 106. С. 1469–1495. DOI: https://doi.org/10.1007/s10994-017-5642-8.
8. Cano A., Krawczyk B. Kappa Updated Ensemble for Drifting Data Stream Mining. Machine Learning. 2020. Vol. 109. С. 175–218. DOI: https://doi.org/10.1007/s10994-019-05840-z.
9. Cano A., Krawczyk B. ROSE: Robust Online Self-Adjusting Ensemble for Continual Learning on Imbalanced Drifting Data Streams. Machine Learning. 2022. Vol. 111. С. 2561–2599. DOI: doi.org/10.1007/s10994-022-06168-x.
10. Mahdi O. A., Pardede E., Ali N., Cao J. Fast Reaction to Sudden Concept Drift in the Absence of Class Labels.
Applied Sciences. 2020. Vol. 10, № 2. Art. 606. DOI: https://doi.org/10.3390/app10020606.
11. Adams J. N., van Zelst S. J., Rose T., van der Aalst W. M. P. Explainable Concept Drift in Process Mining.
Information Systems. 2023. Vol. 114. Art. 102177. DOI: https://doi.org/10.1016/j.is.2023.102177.
12. Ghomeshi H., Gaber M. M., Kovalchuk Y. EACD: Evolutionary Adaptation to Concept Drifts in Data Streams. Data Mining and Knowledge Discovery. 2019. Vol. 33. С. 760–793. DOI: https://doi.org/10.1007/s10618-019-00614-6.
13. Yu E., Lu J., Zhang B., Zhang G. Online Boosting Adaptive Learning under Concept Drift for Multistream Classification. Proceedings of the AAAI Conference on Artificial Intelligence. 2024. Vol. 38, № 15. С. 16522– 16530. DOI: https://doi.org/10.1609/aaai.v38i15.29590.
14. Hu L., Lu Y., Feng Y. Concept Drift Detection Based on Deep Neural Networks and Autoencoders. Applied Sciences. 2025. Vol. 15, № 6. Art. 3056. DOI: https://doi.org/10.3390/app15063056.
References
1. Hinder, F., Vaquet, V., & Hammer, B. (2024). One or two things we know about concept drift – A survey on monitoring in evolving environments. Part A: Detecting concept drift. Frontiers in Artificial Intelligence, 7, Article 1330257. https://doi.org/10.3389/frai.2024.1330257.
2. Li, W., Yang, X., Liu, W., Xia, Y., & Bian, J. (2022). DDG-DA: Data distribution generation for predictable concept drift adaptation. Proceedings of the AAAI Conference on Artificial Intelligence, 36(4), 4092–4100. https://doi.org/10.1609/aaai.v36i4.20327.
3. Baier, L., Schlör, T., Schöffer, J., & Kühl, N. (2021). Detecting concept drift with neural network model uncertainty. arXiv preprint arXiv:2107.01873. https://doi.org/10.48550/arXiv.2107.01873.
4. Alneyadi, S., Sithirasenan, E., & Muthukkumarasamy, V. (2016). A survey on data leakage prevention systems.
Journal of Network and Computer Applications, 62, 137–152. https://doi.org/10.1016/j.jnca.2016.01.008.
5. de Barros, R. S. M., & Santos, S. G. T. C. (2018). A large-scale comparison of concept drift detectors. Information Sciences, 451–452, 348–370. https://doi.org/10.1016/j.ins.2018.04.014.
6. Losing, V., Hammer, B., & Wersing, H. (n.d.). Drift datasets. GitHub repository. Retrieved November 5, 2025, from https://github.com/vlosing/driftDatasets.
7. Gomes, H. M., Bifet, A., Read, J., Barddal, J. P., Enembreck, F., Pfahringer, B., Holmes, G., & Abdessalem, T. (2017). Adaptive random forests for evolving data stream classification. Machine Learning, 106, 1469–1495. https://doi.org/10.1007/s10994-017-5642-8.
8. Cano, A., & Krawczyk, B. (2020). Kappa updated ensemble for drifting data stream mining. Machine Learning, 109, 175–218. https://doi.org/10.1007/s10994-019-05840-z.
9. Cano, A., & Krawczyk, B. (2022). ROSE: Robust online self-adjusting ensemble for continual learning on imbalanced drifting data streams. Machine Learning, 111, 2561–2599. https://doi.org/10.1007/s10994-022-06168-x.
10. Mahdi, O. A., Pardede, E., Ali, N., & Cao, J. (2020). Fast reaction to sudden concept drift in the absence of class labels. Applied Sciences, 10(2), 606. https://doi.org/10.3390/app10020606.
11. Adams, J. N., van Zelst, S. J., Rose, T., & van der Aalst, W. M. P. (2023). Explainable concept drift in process mining. Information Systems, 114, Article 102177. https://doi.org/10.1016/j.is.2023.102177.
12. Ghomeshi, H., Gaber, M. M., & Kovalchuk, Y. (2019). EACD: Evolutionary adaptation to concept drifts in data streams. Data Mining and Knowledge Discovery, 33, 760–793. https://doi.org/10.1007/s10618-019-00614-6.
13. Yu, E., Lu, J., Zhang, B., & Zhang, G. (2024). Online boosting adaptive learning under concept drift for multistream classification. Proceedings of the AAAI Conference on Artificial Intelligence, 38(15), 16522–16530. https://doi.org/10.1609/aaai.v38i15.29590.
14. Hu, L., Lu, Y., & Feng, Y. (2025). Concept drift detection based on deep neural networks and autoencoders.
Applied Sciences, 15(6), Article 3056. https://doi.org/10.3390/app15063056.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Petro Vizhevskyi, Oleg Savenko
This work is licensed under a Creative Commons Attribution 4.0 International License.
