Adaptive multi-paradigm intrusion detection with dynamic decision fusion for robust cybersecurity
Year: 2026
Type: article
Abstract: Cyberattacks are becoming more sophisticated and complex, thereby putting outdated intrusion detection systems (IDSs) to the test, as they are based on a single-detector paradigm and static rule-based evaluation. To overcome these challenges, in this paper, we propose Adaptive Hybrid Intrusion Detection Framework (AHIDF), which combines the signature-based, machine learning-based and deep learning-based detectors in a single framework. A new Adaptive Confidence-Weighted Decision Fusion (ACWDF) mechanism is introduced in the framework, which dynamically weights the detector’s output based on the effectiveness of the detector in recent operations. Moreover, a new Operational Intrusion Detection Readiness Score (OIDRS) is proposed that is used to comprehensively assess detection capability, discriminative performance, false alarm behavior, latency, and the computational overhead. The experimental results showed high accuracy (0.89), precision (0.88), recall (0.86) and F1 score (0.87) in all the categories of cyberattacks and excellent discriminating performance in ROC analysis (AUC up to 0.92). The outcomes show that the proposed adaptive hybrid framework enhances detection robustness and operational readiness by adaptively integrating heterogeneous detection paradigms based on the adaptive decision fusion and deployment-oriented evaluation. The proposed work offers a realistic, scalable approach in today’s cybersecurity environment.
Source: Scientific Reports
Authors: R. Kanimozhi, S. Jagan, K. Lalitha
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,690
OpenAlex ID: W7229069973
