Implementation of CyberCScope.
CyberCScope: Mining Skewed Tensor Streams and Online Anomaly Detection in Cybersecurity Systems.
Kota Nakamura, Koki Kawabata, Shungo Tanaka, Yasuko Matsubara, Yasushi Sakurai.
The Web Conference 2025 short research paper.
CyberCScope is freely available for non-commercial purposes. If you intend to use CyberCScope for a commercial purpose, please contact us by email at [[email protected]]
# Quick demo for partial data of CCI'18
$ sh demo.sh
Pandas.DataFrame
Time + Categorical attributes + (Skewed) Continuous attributes
0| Time | Attribute1 | Attribute2 | Attribute3 | Attribute4 | ...
1| :
2| :
3| :
Note
The method can flexibly handle both categorical and continuous attributes.
It can handle skewness in the continuous attributes.
