Heuristic-Guided Selective Random Forests: Immune-Inspired Heuristic Ensembles for Robust Intrusion Detection in IoT Networks

Abstract

The increasing adoption of the Internet of Things (IoT) has brought considerable benefits to all aspects of daily life, but the IoT remains vulnerable to a wide range of security threats that compromise device continuity and can lead to significant disruption and losses. Traditional intrusion detection systems (IDSs) often rely on machine learning models, which perform relatively well but struggle when faced with unusual traffic patterns. To address these challenges, this study presents a heuristic-based ensemble method. The proposed method combines decision tree detectors with a biologically inspired filtering mechanism and mimics the behavior of the immune system by retaining only trees that achieve a detection accuracy greater than a specified threshold. This method reduces the memory required to store weak decision trees that can affect the final decision, thus improving classification performance. Experimental results demonstrate an overall accuracy of 93.4%, outperforming baseline algorithms. The research presents a framework that balances detection performance with computational efficiency, a critical requirement for IoT networks with limited resources.

Received: 28 September 2025 | Revised: 4 December 2025 | Accepted: 15 December 2025

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The data that support the findings of this study are openly available in the UNSW-NB15 dataset at https://doi.org/10.1109/MilCIS.2015.7348942.

Author Contribution Statement

Saad AL Azzam: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration. Ghassan ALDharhani: Validation, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Supervision, Project administration. Raenu AL Kolandaisamy: Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing, Supervision, Project administration.