Bidirectional Feature-Subset Transfer between Binary Intrusion Detection and Multiclass Attack Attribution in IoT Communication Networks

Ahmed Sabri Ghazi Behadili

Abstract


The set of traffic features used in a reduced binary intrusion detection and multiclass attack attribution may not work equally well in IoT communication networks. This study separates out the impact of granularity of the label selection by moving subsets both ways, keeping the data partitions, classifier settings, and feature budgets constant. The contribution is a paired evaluation of re-use loss complemented with protocol shortcut audits and comparing shared subsets with subsets selected for each task. Four selectors, two tree ensemble classifiers, and 30 matched splits were used for evaluating Edge-IIoTset, the network table and CICIoT2023. Absolute regret is the magnitude of a loss and normalized regret is the loss magnitude relative to the leeway in the goal task. Practically relevant harm was detected in 17, 15 and 22 configurations with the inference with dependence correction on the common grid k = 5-20, respectively. In all of the cases, there was no direction of transfer that was better. Each testbed had a unique shortcut that was associated with MQTT coverage, DNS applicability or a fixed mix of TCP and UDP. They were removed, reducing a few effects but not the primary ones. Recall loss was not always correlated with the rarity of an attack family. Borda fusion tuned on validation data frequently produced a parent subset, and new subsets did not reliably lead to a decrease in loss for the worse task. This validation should therefore be performed on both label views, before shared feature pipelines are used instead of task-specific subsets, against an explicit loss tolerance.

Keywords


feature selection; IoT communication networks; IoT intrusion detection; multiclass classification; network traffic analysis

Full Text:

PDF

References


M. M. Rahman, S. A. Shakil, and M. R. Mustakim, “A Survey on Intrusion Detection System in IoT Networks,” Cyber Security and Applications, Vol. 3, Art. No. 100082, 2025, DOI: 10.1016/j.csa.2024.100082.

M. Fatima, O. Rehman, I. M. H. Rahman, A. Ajmal, and S. J. Park, “Towards Ensemble Feature Selection for Lightweight Intrusion Detection in Resource-Constrained IoT Devices,” Future Internet, Vol. 16, No. 10, Art. No. 368, 2024, DOI: 10.3390/fi16100368.

U. Otokwala, A. Petrovski, and H. Kalutarage, “Optimized Common Features Selection and Deep-Autoencoder (OCFSDA) for Lightweight Intrusion Detection in Internet of Things,” International Journal of Information Security, Vol. 23, No. 4, pp. 2559–2581, 2024, DOI: 10.1007/s10207-024-00855-7.

J. Li, M. S. Othman, H. Chen, and L. M. Yusuf, “Optimizing IoT Intrusion Detection System: Feature Selection versus Feature Extraction in Machine Learning,” Journal of Big Data, Vol. 11, No. 1, Art. No. 36, 2024, DOI: 10.1186/s40537-024-00892-y.

A. Sezgin, M. Ulaş, and A. Boyacı, “Multi-Objective Feature Selection for Intrusion Detection Systems: A Comparative Analysis of Bio-Inspired Optimization Algorithms,” Sensors, Vol. 25, No. 19, Art. No. 6099, 2025, DOI: 10.3390/s25196099.

J. P. Mchina, N. Mduma, and R. S. Sinde, “Adaptive Class-Aware Feature Selection for High-Dimensional and Imbalanced Multi-Class Network Intrusion Detection,” Frontiers in Big Data, Vol. 9, Art. No. 1871346, 2026, DOI: 10.3389/fdata.2026.1871346.

Y. Lu, W. Tian, D. Zhang, and K. Zhao, “Attack Category Mapping Modeling Method based on System-Level Features for Intrusion Detection Systems,” Information Sciences, Vol. 755, Art. No. 123791, 2026, DOI: 10.1016/j.ins.2026.123791.

G. Sakthivel and A. K. Kumarasamy, “A Drift-Aware RS2FS Pipeline with Confidence Gating for IDS,” Scientific Reports, Vol. 16, No. 1, Art. No. 4029, 2026, DOI: 10.1038/s41598-025-34124-x.

R. Sommer and V. Paxson, “Outside the Closed World: on using Machine Learning for Network Intrusion Detection,” in Proc. IEEE Symp. Security and Privacy, 2010, pp. 305–316, DOI: 10.1109/SP.2010.25.

G. Apruzzese, P. Laskov, and J. Schneider, “SoK: Pragmatic Assessment of Machine Learning for Network Intrusion Detection,” in Proc. IEEE 8th European Symp. Security and Privacy (EuroS&P), 2023, pp. 592–614, DOI: 10.1109/EuroSP57164.2023.00042.

R. Flood, G. Engelen, D. Aspinall, and L. Desmet, “Bad Design Smells in Benchmark NIDS Datasets,” in Proc. IEEE 9th European Symp. Security and Privacy (EuroS&P), 2024, pp. 658–675, DOI: 10.1109/EuroSP60621.2024.00042.

L. Liu, G. Engelen, T. Lynar, D. Essam, and W. Joosen, “Error Prevalence in NIDS Datasets: A Case Study on CIC-IDS-2017 and CSE-CIC-IDS-2018,” in Proc. IEEE Conf. Communications and Network Security (CNS), 2022, pp. 254–262, DOI: 10.1109/CNS56114.2022.9947235.

D. Arp et al., “Dos and Don’ts of Machine Learning in Computer Security,” in Proc. 31st USENIX Security Symp. (USENIX Security 22), 2022, pp. 3971–3988. [Online]. Available: https://www.usenix.org/conference/usenixsecurity22/presentation/arp (accessed Oct. 31, 2026).

S. Kaushik et al., “Robust Machine Learning based Intrusion Detection System using Simple Statistical Techniques in Feature Selection,” Scientific Reports, Vol. 15, No. 1, Art. No. 3970, 2025, DOI: 10.1038/s41598-025-88286-9.

M. A. Umar, Z. Chen, K. Shuaib, and Y. Liu, “Effects of Feature Selection and Normalization on Network Intrusion Detection,” Data Science and Management, Vol. 8, No. 1, pp. 23–39, 2025, DOI: 10.1016/j.dsm.2024.08.001.

G. Nassreddine, M. Nassereddine, and O. Al-Khatib, “Ensemble Learning for Network Intrusion Detection based on Correlation and Embedded Feature Selection Techniques,” Computers, Vol. 14, No. 3, Art. No. 82, 2025, DOI: 10.3390/computers14030082.

U. Ahmed et al., “Hybrid Bagging and Boosting with SHAP based Feature Selection for Enhanced Predictive Modeling in Intrusion Detection Systems,” Scientific Reports, Vol. 14, No. 1, Art. No. 30532, 2024, DOI: 10.1038/s41598-024-81151-1.

M. Awad and S. Fraihat, “Recursive Feature Elimination with Cross-Validation with Decision Tree: Feature Selection Method for Machine Learning-based Intrusion Detection Systems,” Journal of Sensor and Actuator Networks, Vol. 12, No. 5, Art. No. 67, 2023, DOI: 10.3390/jsan12050067.

J. Ding, J. Du, H. Wang, and S. Xiao, “A Novel Two-Stage Feature Selection Method based on Random Forest and Improved Genetic Algorithm for Enhancing Classification in Machine Learning,” Scientific Reports, Vol. 15, No. 1, Art. No. 16828, 2025, DOI: 10.1038/s41598-025-01761-1.

Y. K. Saheed and S. Misra, “A Voting Gray Wolf Optimizer-based Ensemble Learning Models for Intrusion Detection in the Internet of Things,” International Journal of Information Security, Vol. 23, No. 3, pp. 1557–1581, 2024, DOI: 10.1007/s10207-023-00803-x.

S. S. Alqahtany, A. Shaikh, and A. Alqazzaz, “Enhanced Grey Wolf Optimization (EGWO) and Random Forest based Mechanism for Intrusion Detection in IoT Networks,” Scientific Reports, Vol. 15, No. 1, Art. No. 1916, 2025, DOI: 10.1038/s41598-024-81147-x.

M. Benmalek and A. Seddiki, “Particle Swarm Optimization-Enhanced Machine Learning and Deep Learning Techniques for Internet of Things Intrusion Detection,” Data Science and Management, Vol. 8, No. 4, pp. 423–435, 2025, DOI: 10.1016/j.dsm.2025.02.005.

A. K. Dey, G. P. Gupta, and S. P. Sahu, “A Metaheuristic-based Ensemble Feature Selection Framework for Cyber Threat Detection in IoT-Enabled Networks,” Decision Analytics Journal, Vol. 7, Art. No. 100206, 2023, DOI: 10.1016/j.dajour.2023.100206.

S.-M. Tseng, Y.-Q. Wang, and Y.-C. Wang, “Multi-Class Intrusion Detection based on Transformer for IoT Networks using CIC-IoT-2023 Dataset,” Future Internet, Vol. 16, No. 8, Art. No. 284, 2024, DOI: 10.3390/fi16080284.

C. Christy, A. Nirmala, A. M. O. Teena, and A. I. Amali, “Machine Learning based Multi-Stage Intrusion Detection System and Feature Selection Ensemble Security in Cloud Assisted Vehicular Ad Hoc Networks,” Scientific Reports, Vol. 15, No. 1, Art. No. 27058, 2025, DOI: 10.1038/s41598-025-96303-0.

M. N. Amr, A. Elliethy, T. Mekkawy, and A. Mahran, “Sequential Submodular Feature-Sample Selection for Lightweight and Update-Efficient IoT Intrusion Detection,” Scientific Reports, Vol. 16, No. 1, Art. No. 26315, 2026, DOI: 10.1038/s41598-026-66614-x.

M. B. Musthafa et al., “Optimizing IoT Intrusion Detection using Balanced Class Distribution, Feature Selection, and Ensemble Machine Learning Techniques,” Sensors, Vol. 24, No. 13, Art. No. 4293, 2024, DOI: 10.3390/s24134293.

C. Kılıç and G. Şengül, “SHAP-Guided Feature Selection for Cross-Dataset Generalization in Network Intrusion Detection Systems,” IEEE Access, Vol. 14, pp. 90926–90936, 2026, DOI: 10.1109/ACCESS.2026.3703481.

M. Sarhan, S. Layeghy, and M. Portmann, “Evaluating Standard Feature Sets Towards Increased Generalisability and Explainability of ML-based Network Intrusion Detection,” Big Data Research, Vol. 30, Art. No. 100359, 2022, DOI: 10.1016/j.bdr.2022.100359.

M. Wang, N. Yang, and N. Weng, “K-GetNID: Knowledge-Guided Graphs for Early and Transferable Network Intrusion Detection,” IEEE Transactions on Information Forensics and Security, Vol. 19, pp. 7147–7160, 2024, DOI: 10.1109/TIFS.2024.3431932.

M. M. Islam, U. Salsabil, M. Nurmamatov, and S. Hossain, “Lightweight Intrusion Detection Systems for IoT–Edge Environments: A PRISMA-ScR Systematic Review of Deployability Evidence and a Unified Assessment Framework,” Future Internet, Vol. 18, No. 6, Art. No. 300, 2026, DOI: 10.3390/fi18060300.

M. A. Ferrag, O. Friha, D. Hamouda, L. Maglaras, and H. Janicke, “Edge-IIoTset: A New Comprehensive Realistic Cyber Security Dataset of IoT and IIoT Applications for Centralized and Federated Learning,” IEEE Access, Vol. 10, pp. 40281–40306, 2022, DOI: 10.1109/ACCESS.2022.3165809.

N. Moustafa, “A New Distributed Architecture for Evaluating AI-based Security Systems at the Edge: Network TON_IoT Datasets,” Sustainable Cities and Society, Vol. 72, Art. No. 102994, 2021, DOI: 10.1016/j.scs.2021.102994.

E. C. P. Neto, S. Dadkhah, R. Ferreira, A. Zohourian, R. Lu, and A. A. Ghorbani, “CICIoT2023: A Real-Time Dataset and Benchmark for Large-Scale Attacks in IoT Environment,” Sensors, Vol. 23, No. 13, Art. No. 5941, 2023, DOI: 10.3390/s23135941.

L. Breiman, “Random forests,” Machine Learning, Vol. 45, No. 1, pp. 5–32, 2001, DOI: 10.1023/A:1010933404324.

G. Ke et al., “LightGBM: A Highly Efficient Gradient Boosting Decision Tree,” in Advances in Neural Information Processing Systems, Vol. 30, 2017. [Online]. Available: https://proceedings.neurips.cc/paper_files/paper/2017/hash/6449f44a102fde848669bdd9eb6b76fa-Abstract.html (accessed Oct. 31, 2026).

F. Wilcoxon, “Individual Comparisons by Ranking Methods,” Biometrics Bulletin, Vol. 1, No. 6, pp. 80–83, 1945, DOI: 10.2307/3001968.

J. Demšar, “Statistical Comparisons of Classifiers Over Multiple Data Sets,” Journal of Machine Learning Research, Vol. 7, pp. 1–30, 2006. [Online]. Available: https://www.jmlr.org/papers/v7/demsar06a.html (accessed Oct. 31, 2026).

C. Nadeau and Y. Bengio, “Inference for the Generalization Error,” Machine Learning, Vol. 52, No. 3, pp. 239–281, 2003, DOI: 10.1023/A:1024068626366.

S. Holm, “A Simple Sequentially Rejective Multiple Test Procedure,” Scandinavian Journal of Statistics, Vol. 6, No. 2, pp. 65–70, 1979, DOI: 10.2307/4615733. [Online]. Available: https://www.jstor.org/stable/4615733 (accessed Oct. 31, 2026).




DOI: https://doi.org/10.32520/stmsi.v15i11.7048

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.