Dear all,
I am looking for intrusion detection datasets that are unbalanced (number of normal connections higher than number of attacks). So far I have only found CICIDS2017 and CICIDS2018 with these characteristics. Can you suggest me some other datasets? Possibly with labels and in csv format?
Kindly, have a look at our works and dataset:
https://www.kaggle.com/azalhowaide/iot-dataset-for-intrusion-detection-systems-ids/settings
Alaa Alhowaide thanks for your reply. What is the mean of the label of the dataset? 1 are attacks samples and 0 are normal samples?
Giuseppina Andresini 0 stands for attack and 1 stands for normal samples. Next week I will add a version with a categorical class label. I updated the dataset description as well. Thanks a lot for notifying me regarding this point.
Giuseppina Andresini A new version of the dataset is added. The new version includes all the records and three categorical features representing the devices' names, attacks, and sub-attacks names.
you can use this dataset
(PDF) X-IIoTID: A Connectivity-and Device-agnostic Intrusion Dataset for Industrial Internet of Things (researchgate.net)
- Badges
- Science topic
- Similar topics
- Biological Science
- Bioecology
- Systems Ecology
- Landscape Ecology
- Connectivity