Unified Heterogeneous Intrusion Detection Dataset
收藏Zenodo2026-03-03 更新2026-05-26 收录
下载链接：
https://zenodo.org/doi/10.5281/zenodo.18660929
下载链接
链接失效反馈官方服务：
资源简介：
This dataset provides a unified heterogeneous intrusion detection testbed constructed by reprocessing raw network traffic from multiple widely used benchmark datasets. The dataset was developed to support reproducible evaluation of Intrusion Detection System (IDS) pipelines under heterogeneous traffic conditions while preserving each dataset’s native feature extraction methodology.
Features were re-extracted directly from raw Packet Capture (PCAP) files using the same tools and methodologies employed by the original dataset authors. Extracted traffic instances were pooled into a unified heterogeneous traffic testbed. Traffic was labeled and mapped into a consistent binary classification scheme. This approach ensures faithful reproduction of dataset-specific feature representations while enabling unified evaluation under a shared heterogeneous traffic distribution.
The dataset aggregates traffic derived from the following publicly available intrusion detection benchmarks:
KDD’99 (derived from DARPA’98 tcpdump PCAP traces)
NSL-KDD (derived from DARPA’98 tcpdump PCAP traces)
UNSW-NB15 (features extracted using Argus and Zeek)
CIC-IDS2017 (features extracted using CICFlowMeter-V2)
CSE-CIC-IDS2018 (features extracted using CICFlowMeter-V2)
提供机构：
Zenodo创建时间：
2026-02-16
