Incremental Feature Learning for Cybersecurity under Evolving Feature Spaces

• Rafael S. Lemos UFC
• Davyson S. Ribeiro UFC
• Francisco R. P. da Ponte UFC
• César Lincoln C. Mattos UFC
• Emanuel B. Rodrigues UFC

DOI: https://doi.org/10.5753/sbseg.2026.26941

Resumo

Cybersecurity systems continuously evolve by incorporating new data attributes, requiring machine learning models to adapt without costly retraining. This work proposes two Incremental Feature Learning (IFL) approaches for intrusion detection and vulnerability risk classification: a hybrid Naive Bayes model and a Multilayer Perceptron with partial weight freezing. Experimental results show that both approaches effectively adapt to progressively expanding feature spaces while maintaining competitive computational efficiency. The proposed MLP achieved over 93% accuracy for intrusion detection, whereas the incremental strategies reduced learning cost and memory consumption without full model retraining.

Referências

Aggarwal, C. (2020). Data Classification: Algorithms and Applications. Chapman & Hall/CRC Data Mining and Knowledge Discovery. CRC Press. ISBN: 9780367659141. URL: [link].

Appasani, D., C. S. Bokkisam, and S. Surendran (2024). “An Incremental Naive Bayes Learner for Real-time Health Prediction”. In: Procedia Computer Science 235, pp. 2942–2954.

Dissanayake, N. et al. (2022). “Software security patch management-A systematic literature review of challenges, approaches, tools and practices”. In: Information and Software Technology 144, p. 106771.

Hozouri, A., A. Mirzaei, and M. Effatparvar (2025). “A comprehensive survey on intrusion detection systems with advances in machine learning, deep learning and emerging cybersecurity challenges”. In: Discover Artificial Intelligence 5.1, p. 314.

Kim, C. H. et al. (2024). “Robust and Adaptive Incremental Learning for Varying Feature Space”. In: IEEE Access 12, pp. 64177–64192.

Kuang, Z. et al. (2024). “Incremental attribute learning by knowledge distillation method”. In: Journal of Computational Design and Engineering 11.5, pp. 259–283.

Müller, A. C. and S. Guido (2016). Introduction to machine learning with Python: a guide for data scientists. ” O’Reilly Media, Inc.”.

Pinto, A. et al. (2023). “Survey on intrusion detection systems based on machine learning techniques for the protection of critical infrastructure”. In: Sensors 23.5, p. 2415.

Ponte, F. R. P. da, E. B. Rodrigues, and C. L. C. Mattos (2023). “CVEjoin: An Information Security Vulnerability and Threat Intelligence Dataset”. In: International Conference on Advanced Information Networking and Applications. Springer, pp. 380–392.

Sadreddin, A. and S. Sadaoui (2022). “Incremental Feature Learning for Fraud Data Stream”. In: Proceedings of the 14th International Conference on Agents and Artificial Intelligence (ICAART), pp. 268–275.

Sarhan, M., S. Layeghy, and M. Portmann (2022). “Towards a standard feature set for network intrusion detection system datasets”. In: Mobile networks and applications 27.1, pp. 357–370.

Secretaria de Governo Digital (2023). Guia de Gerenciamento de Vulnerabilidades. Versão 2.0. URL: [link].

Sharafaldin, I., A. H. Lashkari, and A. A. Ghorbani (2018). “Toward Generating a New Intrusion Detection Dataset and Intrusion Traffic Characterization”. In: Proceedings of the 4th International Conference on Information Systems Security and Privacy (ICISSP), pp. 108–116.

Van de Ven, G. M., T. Tuytelaars, and A. S. Tolias (2022). “Three types of incremental learning”. In: Nature Machine Intelligence 4.12, pp. 1185–1197.

Yang, Z. et al. (2023). “A new three-way incremental naive bayes classifier”. In: Electronics 12.7, p. 1730.