Tag: CIC-IDS2017

A new integrated framework combining quantum-inspired optimization and deep learning achieves high accuracy in detecting diverse cyberattacks across major cloud ...

Researchers developed ZeroDayDetectAI, an unsupervised ensemble framework combining autoencoders and latent-space clustering that detected unseen cyberattacks with 96.78 percent accuracy ...

A study evaluating SHAP and LIME across network intrusion detection classifiers demonstrates that post-hoc explanation agreement remains limited and does ...

A controlled audit of conformal prediction on two intrusion detection benchmarks shows that marginal coverage can hide rare-class failures as ...

A new explainable AI framework for network intrusion detection audited its semantic rule layer on held-out benchmark data, reporting high ...

A new class-incremental learning framework called DARE-IDS uses diversity-aware replay exemplars to help intrusion detection systems learn new attack classes ...