What question did this study set out to answer?

To develop an interpretable network intrusion detection system that addresses challenges in high-dimensional data and class imbalance.

March 30, 2026Open Access

Interpretable Intrusion Detection with TabNet Attention Masks Enhanced by Information Gain and Grey Wolf Optimization

Puntos clave

To develop an interpretable network intrusion detection system that addresses challenges in high-dimensional data and class imbalance.
Utilized Information Gain for feature selection to reduce dimensionality.
Employed Grey Wolf Optimizer for hyperparameter optimization of TabNet.
Applied TabNet's attention mechanism for interpretable predictions.
Evaluated the framework on four benchmark datasets.
Achieved 99.47 ± 0.11% accuracy on the CIC-IDS2017 dataset.
Outperformed the second-best method by 0.35%, indicating significant improvement.
Demonstrated high statistical significance with p < 0.001.

Resumen

Network intrusion detection systems (NIDS) are critical for protecting modern cyber infrastructure against evolving threats. Traditional machine learning approaches often struggle with high-dimensional network traffic data and class imbalance. This paper proposes a novel hybrid intrusion detection framework combining Information Gain (IG) for feature selection, Grey Wolf Optimizer (GWO) for hyperparameter optimization, and TabNet for deep learning-based classification. The IG module reduces dimensionality by selecting the most discriminative features, GWO optimizes TabNet’s architecture and learning parameters, and TabNet’s attention-based mechanism enables interpretable predictions. We evaluate our approach on four benchmark datasets: CIC-IDS2017, NSL-KDD, UNSW-NB15, and CIC-DDoS2019. On CIC-IDS2017, our method achieves 99.47 ± 0.11% accuracy, outperforming the second-best (GWO-TabNet:99.12 ± 0.15%) by 0.35% with high statistical significance (p < 0.001).

Leer artículo completoexternamente

Me gusta

Guardar

Ver artículo completo