An Ensemble-Based Network Intrusion Detection System Using Random Forest Algorithm: A 2025 Perspective

With the ever-evolving threat landscape in cyberspace, traditional intrusion detection systems (IDS) struggle to maintain efficacy against novel and complex attacks. This paper presents an ensemble-based Network Intrusion Detection System (NIDS) employing the Random Forest algorithm, evaluated using the UNSW-NB15 dataset. The ensemble approach combines the strengths of multiple base classifiers—K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Decision Tree (DT)—with Random Forest to achieve robust and high-performing detection capabilities. Experimental results demonstrate improved accuracy, precision, recall, and F1-score compared to individual classifiers.