CoolFace
Datasetpublic

lacg030175/CIC-IoT-2023-full

CIC-IoT-2023 Full Dataset (46M+ rows) The FULL CICIoT2023 dataset — all 38,508,041 rows, no subsampling. Configuration random_3way (default) — 80/10/10 Three-Way Split Stratified random split with fully separated sets: Train (80%): 30,806,432 rows — model training and architecture search Test (10%): 3,850,804 rows — threshold calibration (held out from training) Validation (10%): 3,850,805 rows — final reported metrics (never touched) from… See the full description on the dataset page: https://huggingface.co/datasets/lacg030175/CIC-IoT-2023-full.

sourceHugging Facecc-by-4.0updated 6mo agoView on Hugging Face
0likes79downloads
Dataset Card

CIC-IoT-2023 Full Dataset (46M+ rows)

The FULL CICIoT2023 dataset — all 38,508,041 rows, no subsampling.

Configuration

random_3way (default) — 80/10/10 Three-Way Split

Stratified random split with fully separated sets:

  • —Train (80%): 30,806,432 rows — model training and architecture search
  • —Test (10%): 3,850,804 rows — threshold calibration (held out from training)
  • —Validation (10%): 3,850,805 rows — final reported metrics (never touched)
python
from datasets import load_dataset
ds = load_dataset("lacg030175/CIC-IoT-2023-full", "random_3way")

Class Distribution

ClassCount%
Benign1,098,1262.9%
Attack37,409,91597.1%

Comparison with Subsampled Version

VersionRowsBenign %Purpose
lacg030175/CIC-IoT-2023~1.3M~15%Fast experiments (112 runs)
This dataset38,508,0412.9%Literature comparison

Top-20 RF Features

  1. 1.Number
  2. 2.ackflagnumber
  3. 3.Header_Length
  4. 4.HTTPS
  5. 5.TimeToLive
  6. 6.pshflagnumber
  7. 7.Rate
  8. 8.IAT
  9. 9.Max
  10. 10.ack_count
  11. 11.Tot sum
  12. 12.Variance
  13. 13.Tot size
  14. 14.Std
  15. 15.Min
  16. 16.AVG
  17. 17.syn_count
  18. 18.synflagnumber
  19. 19.HTTP
  20. 20.fin_count

Attack Types (7 classes, 33 sub-types)

ClassSub-types
BenignBenignTraffic
BruteForceDictionaryBruteForce
DDoS12 sub-types
DoS4 sub-types
Mirai3 sub-types
Recon5 sub-types
Spoofing2 sub-types
Web-based6 sub-types

Labels

  • —Binary (label): 0 = Benign, 1 = Attack
  • —Multi-class (Label): 34 categories
  • —Grouped (attack_class): 8 classes

Citation

bibtex
@article{neto2023ciciot,
  title={CICIoT2023: A Real-Time Dataset and Benchmark for Large-Scale Attacks in IoT Environment},
  author={Neto, Euclides Carlos Pinto and others},
  journal={Sensors},
  volume={23},
  number={13},
  year={2023},
  publisher={MDPI}
}

License

CC BY 4.0 — original dataset by the Canadian Institute for Cybersecurity, University of New Brunswick.