A Review of Ensemble Learning Techniques and Public Dataset for Intrusion Detection Systems
Fifto Nugroho, Ema Utami, Kusrini Kusrini, Arief Setyanto · 2025
As cyber threats grow increasingly complex, advanced Intrusion Detection Systems (IDS) development has become a critical focus. Ensemble learning methods have shown great potential in enhancing IDS performance by improving accuracy, increasing robustness, and minimizing false positive rates. IDS public datasets are widely used as benchmarks, providing diverse attack scenarios and realistic traffic patterns for evaluating these techniques. This paper reviews advancements in ensemble learning methods for IDS, highlighting their effectiveness in addressing key challenges such as data imbalance and computational complexity. Additionally, it identifies the limitations of current approaches and proposes future research directions, including real-time IDS optimization and the development of explainable ensemble learning models.