Abstract

This paper presents a new hybrid machine learning-based framework for robust anomaly detection, specifically for critical network environments. This method is based on the implementation of an SVM classifier and Whale Optimization Algorithm (WOA), a bio-inspired metaheuristic technique, aimed at resolving the crucial bottleneck of classifier hyperparameter optimization. The main objective is to achieve a remarkable increase in classification accuracy and decrease the false positive rate (FPR), which are significant shortcomings in current security systems. WOA is used to optimize the critical parameters of SVM efficiently so as to optimize its abilities for discriminating between normal and malicious network activities. The SWaT (Secure Water Treatment) dataset, a high-fidelity benchmark for industrial control systems, was selected for the experimental evaluation. As a consequence of the results, the accuracy of the optimized WOA-SVM model reaches 95.3% and the false positive rate reduces to 3.6%. Our results validate how the optimization-enhanced implementation provides a more robust and accurate security mechanism for contemporary network infrastructures

Keywords

  • Hybrid Machine Learning
  • Whale Optimization Algorithm
  • Support Vector Machine
  • Anomaly Detection
  • Cy

References

  1. L. Shan et al., "IoT Network intrusion detection system using optimization-based hybrid machine learning," Journal of Network and Computer Applications, vol. 250, 2025.
  2. F. Alhayan et al., "Design of advanced intrusion detection in cybersecurity using ensemble learning with improved Beluga Whale Optimization (IDCS-ELIBWO) technique," Ain Shams Engineering Journal, vol. 16, no. 6, 2025.
  3. D. Wang et al., "Research on the Detection of Network Intrusion Prevention System Based on Improved WOA-SVM," Informatica, vol. 44, no. 4, 2020.
  4. Q. M. Alzubi et al., "Intrusion detection system based on hybridizing a modified binary Grey Wolf Optimization and Particle Swarm Optimization," Expert Systems with Applications, vol. 209, 2022.
  5. D. J. Kalita et al., "A novel adaptive optimization framework for SVM hyper-parameter tuning in intrusion detection," Expert Systems with Applications, vol. 211, 2023.
  6. M. Habib et al., "Bio-inspired optimization of feature selection and SVM hyper-parameters for enhanced intrusion detection," Applied Soft Computing, vol. 161, 2025.
  7. S. Jaradat et al., "Cyberattack detection on SWaT plant industrial control system using deep learning," Journal of Electrical Systems and Information Technology, vol. 11, no. 1, 2024.
  8. Y. A. Ali et al., "Hyperparameter Search for Machine Learning Algorithms Using Metaheuristic Optimization: A Comparative Study," Applied Sciences, vol. 11, no. 2, 2023.
  9. D. A. Al-Qudah et al., "A Harris Hawks Optimized SVM Framework for Securing IoT Networks," Wireless Personal Communications, vol. 136, 2025.
  10. D. A. Al-Qudah et al., "A Harris Hawks Optimized SVM Framework for Securing IoT Networks," Wireless Personal Communications, vol. 136, 2025.
  11. D. D. Nguyen et al., "Improving intrusion detection in SCADA systems using stacking ensemble of tree-based models," Bulletin of Electrical Engineering and Informatics, vol. 11, no. 6, 2022.
  12. O. A. Alimi et al., "Supervised learning based intrusion detection for SCADA systems," IEEE Access, vol. 10, 2022.
  13. M. Altaha et al., "An Autoencoder-Based Network Intrusion Detection System for the SCADA System," Journal of Communications, vol. 16, no. 6, pp. 504-511, 2021.
  14. A. Kumar et al., "A review on metaheuristic algorithms for intrusion detection systems," Journal of Ambient Intelligence and Humanized Computing, vol. 12, 2021.
  15. I. A. Khan, D. Pi, Z. U. Khan, Y. Hussain, and A. Nawaz, "HML-IDS: A hybrid-multilevel anomaly prediction approach for intrusion detection in SCADA systems," IEEE Access, vol. 7, 2019.