Hybrid Ensemble Learning for Network Intrusion Detection: Integrating Gradient Boosting and Deep Neural Networks

Ammar Adel Aladel

Abstract


Network intrusion detection systems (NIDS) are critical for safeguarding digital infrastructure against evolving cyber threats. AI technologies like machine learning and deep learning have been effective in detecting malicious traffic, but they each come with their own advantages and disadvantages. In this paper, we present a hybrid stacking ensemble model to detect network attacks, suspicious activities, and normal traffic in UGRansome dataset, combining three highly efficient gradient boosting algorithms, namely XGBoost, LightGBM, and CatBoost by integrating with multi-layer deep neural network (MLP). We have employed the following preprocessing techniques: Logarithmic transformation, square-root transformation, Yeo–Johnson transformation, standardisation and label encoding on categorical features in our preprocessing pipeline. A logistic regression model, called the stacking meta‐classifier, is used to learn how to combine the probabilistic output of base estimators. The performance of the proposed ensemble is assessed through extensive experiments with an 80/20 train-test split, which shows that the proposed ensemble achieves an accuracy of 99.5% and a macro-averaged F1-Score of 99.5%, outperforming each individual model. Perhaps most impressively, LightGBM is able to achieve the same accuracy as the ensemble, but the ensemble is more robust with lower misclassification on all classes and higher precision and recall. The results of this work demonstrate the effectiveness of hybrid stacking of multiple classification decision boundaries to effectively exploit the complementary characteristics, yielding a reliable and deployable approach to real‑time network threat classification.

Keywords


anomaly detection; ensemble learning; hybrid stacking; intrusion detection; network security; UGRansome dataset

Full Text:

PDF

References


N. Singh, R. Buyya, and H. Kim, “Securing Cloud-based Internet of Things: Challenges and Mitigations,” Sensors, Vol. 25, No. 1, 2025.

H. Xu, W. Yu, D. Griffith, and N. Golmie, “A Survey on Industrial Internet of Things: A Cyber-Physical Systems Perspective.,” IEEE access Pract. Innov. open Solut., Vol. 6, 2018.

U. Ahmed et al., “Signature-based Intrusion Detection using Machine Learning and Deep Learning Approaches Empowered with Fuzzy Clustering.,” SCI. Rep., Vol. 15, No. 1, p. 1726, Jan. 2025.

J. Mao, X. Yang, B. Hu, Y. Lu, and G. Yin, “Intrusion Detection System based on Multi-Level Feature Extraction and Inductive Network,” Electronics, Vol. 14, No. 1, 2025.

K. C. Mouli, B. Indupriya, D. Ushasree, D. C. V Raghavendran, B. Rawat, and B. Madhu, “Network Intrusion Detection using ML Techniques for Sustainable Information System,” E3S Web Conf., Vol. 430, Oct. 2023.

J. M. Ahn, J. Kim, and K. Kim, “Ensemble Machine Learning of Gradient Boosting (XGBoost, LightGBM, CatBoost) and Attention-based CNN-LSTM for Harmful Algal Blooms Forecasting.,” Toxins (Basel)., Vol. 15, No. 10, Oct. 2023.

Ali, M. L, K. Thakur, S. Schmeelk, J. Debello, and D. Dragos, “Deep Learning vs. Machine Learning for Intrusion Detection in Computer Networks: A Comparative Study,” Applied Sciences, Vol. 15, No. 4. 2025.

P. Koukaras and C. Tjortjis, “Data Preprocessing and Feature Engineering for Data Mining: Techniques, Tools, and Best Practices,” AI, Vol. 6, No. 10, 2025.

K.-L. Du, R. Zhang, B. Jiang, J. Zeng, and J. Lu, “Foundations and Innovations in Data Fusion and Ensemble Learning for Effective Consensus,” Mathematics, Vol. 13, No. 4, 2025.

I. Chaudhary and S. Adhikari, “Ransomware Detection using Machine Learning Techniques,” Res. CAB A J. Res. Dev., Vol. 3, pp. 96–114, Aug. 2024.

P. Azugo, H. Venter, and M. W. Nkongolo, “Ransomware Detection and Classification using Random Forest: A Case Study with the UGRansome2024 Dataset,” 2024.

S. J. Nhlapo and M. W. Nkongolo, “Zero-Day Attack and Ransomware Detection,” ArXiv, pp. 1–14, 2024.

M. Nkongolo, J. P. van Deventer, and S. M. Kasongo, “UGRansome1819: A Novel Dataset for Anomaly Detection and Zero-Day Threats,” Information, Vol. 12, No. 10, 2021.

Z. Ji, J. Gong, and J. Feng, “A Novel Deep Learning Approach for Anomaly Detection of Time Series Data,” SCI. Program., Vol. 2021, Jul. 2021.

A. A. Alhashmi, A. A. Darem, A. B. Alshammari, L. A. Darem, H. K. Sheatah, and R. Effghi, “Ransomware Early Detection Techniques,” Eng. Technol. Appl. SCI. Res., Vol. 14, No. 3, pp. 14497–14503, 2024.

P. Yan, T. T. Khoei, R. S. Hyder, and R. S. Hyder, “A Dual-Stage Ensemble Approach to Detect and Classify Ransomware Attacks,” in 2024 IEEE 15th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), 2024, pp. 781–786.

M. Wa Nkongolo, J. van Deventer, and K. Mambwe Sydney, “UGRansome1819: A Novel Dataset for Anomaly Detection and Zero-Day Threats,” Information, Vol. 12, p. 32, Sep. 2021.

M. W. Nkongolo, “Zero-Day Vulnerability Prevention with Recursive Feature Elimination and Ensemble Learning,” IACR Cryptol. ePrint Arch., Vol. 2023, p. 1843, 2023.

P. C. Karanam and S. Srinivasan, “Enhancement of Cybersecurity with Machine Learning: Applying Ensemble Classification and Anomaly Detection Approaches to Find Zero-Day Attacks in Network Traffic,” in 2025 5th International Conference on Emerging Research in Electronics, Computer Science and Technology (ICERECT), 2025, pp. 1–6.




DOI: https://doi.org/10.32520/stmsi.v15i11.7006

Article Metrics

Abstract view : 0 times
PDF - 0 times

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.