Attack types and descriptions in CICIDS datasets.

<div><p>Deep learning methods have achieved remarkable progress in network intrusion detection. However, their performance often deteriorates significantly in real-world scenarios characterized by limited attack samples and substantial domain shifts. To address this challenge, we propose...

Full description

Saved in:

Bibliographic Details
Main Author:	Congyuan Xu (3338283) (author)
Other Authors:	Donghui Li (94039) (author), Zihao Liu (4020458) (author), Jun Yang (2084) (author), Qinfeng Shen (21647973) (author), Ningbing Tong (21647976) (author)
Published:	2025
Subjects:	Cell Biology Genetics Space Science Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified two benchmark datasets obtained via two modified mamba architecture improve feature discriminability branch feature extractor achieved remarkable progress substantial domain shifts network intrusion detection limited attack samples jointly capture spatial frequency domain features frequency domain characteristics domain feature fusion world scenarios characterized attention </ p domain bidirectional cross domain scenarios domain fusion frequency domains domain generalization query samples network traffic capture long bidirectional cross attention module attention mechanism traffic sequences thereby proving temporal patterns spectral structure shot setting shot conditions semantic gap range dependencies practical applicability outperforming state method adopts extensive experiments enables deep dct ), art methods
Tags:	Add Tag No Tags, Be the first to tag this record!

Attack types and descriptions in CICIDS datasets.

Similar Items