Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM66770
Volume:
10 |
8 |
Aug 2026
Pages:
1-10
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Computer Science and Engineering
All Saints' College of Technology
Computer Science and Engineering
Adwaita Mission Institute of Technology
Computer Science and Engineering
Adwaita Mission Institute of Technology
Mechanical Engineering
Adwaita Mission Institute of Technology
Mechanical Engineering
Adwaita Mission Institute of Technology
The rapid proliferation of the Internet of Things (IoT) has enabled seamless connectivity among billions of smart devices across diverse application domains, including healthcare, industrial automation, smart cities, transportation, and precision agriculture. While this interconnected ecosystem offers significant operational and economic benefits, it also expands the attack surface, making IoT networks increasingly susceptible to cyber threats such as Distributed Denial-of-Service (DDoS), botnet, brute-force, reconnaissance, and spoofing attacks [1]. Conventional intrusion detection systems (IDSs), primarily based on signature matching and predefined rules, exhibit limited capability in identifying unknown or evolving attack patterns, particularly within dynamic and resource-constrained IoT environments [2]. Machine learning (ML)-based IDSs have demonstrated improved adaptability; however, models relying on a single classifier often suffer from limited generalisation, high false alarm rates, and inconsistent detection performance across heterogeneous traffic patterns. This paper proposes an AI-Driven Intrusion Detection System (AI-IDS) based on an Adaptive Confidence-Based Dynamic Weighted Ensemble (ACDWE) framework to enhance cyber attack detection in IoT networks. The proposed framework integrates multiple supervised machine learning classifiers through a dynamic confidence-weighting mechanism that assigns adaptive importance to each classifier according to its predictive reliability. In addition, a hybrid feature optimization strategy is employed to eliminate redundant network traffic attributes, thereby improving computational efficiency while preserving discriminative information. The framework is designed to support real-time intrusion detection with enhanced robustness against diverse attack categories. The proposed model will be evaluated using publicly available benchmark IoT intrusion datasets, including CICIoT2023, Bot-IoT, and UNSW-NB15, and assessed using standard performance metrics such as Accuracy, Precision, Recall, F1-score, Receiver Operating Characteristic–Area Under Curve (ROC-AUC), Matthews Correlation Coefficient (MCC), Detection Rate (DR), and False Alarm Rate (FAR). Comparative analysis with conventional machine learning models and existing ensemble-based IDS approaches is expected to demonstrate improved detection accuracy, lower false positive rates, and better generalization across multiple attack scenarios [3]. The proposed framework aims to provide a scalable, adaptive, and computationally efficient intrusion detection solution suitable for next-generation IoT security applications.
KEYWORDS
This paper presented an AI-Driven Intrusion Detection System (AI-IDS) designed for resource-constrained, dynamic Internet of Things (IoT) environments using an Adaptive Confidence-Based Dynamic Weighted Ensemble (ACDWE) framework. The proposed framework integrates four supervised tree-based ensemble models (Random Forest, XGBoost, LightGBM, and Extra Trees) with dynamic confidence-weighted decision aggregation.A Hybrid Feature Selection (HFS) strategy eliminated over 60% of redundant attributes, enhancing classification accuracy while significantly lowering inference latency.Validation across three benchmark datasets (CICIoT2023, Bot-IoT, and UNSW-NB15) demonstrated superior overall performance (98.45% average accuracy, 98.10% Detection Rate, and 0.85% False Alarm Rate) compared to individual base classifiers and static voting systems.
A. Hassanjani, M. S. Ebrahimzadeh, and H. R. Naji, "IoT security threats and defensive approaches: A systematic literature review," IEEE Access, vol. 10, pp. 84120–84145, 2022.N. Moustafa, B. Turnbull, and K. K. R. Choo, "An ensemble intrusion detection technique based on proposed statistical flow features for protecting network infrastructure," IEEE Transactions on Network and Service Management, vol. 16, no. 2, pp. 483–496, 2019.M. A. Ferrag, L. Maglaras, S. Moschoyiannis, and H. Janicke, "Deep learning for cyber security intrusion detection: Approaches, datasets, and comparative study," Journal of Information Security and Applications, vol. 50, p. 102419, 2020.Z. K. Maseer, R. Yusof, S. A. Mostafa, N. Bahaman, O. Musa, and B. A. S. Al-rimy, "Benchmarking radar-based network intrusion detection systems using machine learning and deep learning," IEEE Access, vol. 9, pp. 22351–22370, 2021.I. Ahmad, M. Basheri, A. Iqbal, and A. Rahim, "Performance comparison of support vector machine, random forest, and XGBoost on network intrusion detection datasets," IEEE Access, vol. 9, pp. 165522–165531, 2021.A. Khraisat, A. Alazab, and R. Singh, "A ensemble learning-based intrusion detection framework for the Internet of Things," IEEE Internet of Things Journal, vol. 9, no. 18, pp. 17211–17220, 2022.S. M. Kasongo and Y. Sun, "A machine learning method with filter-based feature selection for intrusion detection systems," Security and Communication Networks, vol. 2020, pp. 1–12, 2020.M. Z. Al-Faiz and K. A. Farhan, "Feature reduction using PCA and Chi-Square for IoT intrusion detection systems," IEEE Systems Journal, vol. 15, no. 3, pp. 4112–4121, 2021.S. T. Zargar, J. Joshi, and D. Tipper, "A survey of defense mechanisms against distributed denial of service (DDoS) flooding attacks," IEEE Communications Surveys & Tutorials, vol. 15, no. 4, pp. 2046–2069, 2013.S. Garcia, M. Grill, J. Stiborek, and A. Zunino, "An empirical comparison of botnet detection methods," Computers & Security, vol. 45, pp. 100–123, 2014.
Publication Date
8/18/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
3
Total views since publication
PDF Downloads
3
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...