CICDDoS2019 — Exploitative Attack Data Preprocessing & Analysis

A reproducible notebook for preprocessing exploitative attack data from the CICDDoS2019 dataset.
This repo streamlines how I handle, clean, and structure this dataset for downstream ML/DL experiments on DDoS detection.

Highlights

• Loads CICDDoS2019 exploitative attack flows
• Cleans feature set, encodes categorical variables
• Balances class distribution with stratified splits
• Provides baseline EDA plots and class stats
• Exports clean CSV for training anomaly detection models

Notebook: notebooks/Preprocessing_Exploitative_Attack_Data_from_the_CICDDOS2019_Dataset.ipynb

Dataset

• Source: CICDDoS2019 (Canadian Institute for Cybersecurity)
• Download the original dataset from CIC and place CSVs in data/
• Update notebook paths if needed

Environment (Python)

python -m venv .venv
source .venv/bin/activate   # Windows: .venv\Scripts\activate
pip install -r requirements.txt
# Launch Jupyter
pip install notebook
jupyter notebook

Requirements

• pandas
• numpy
• scikit-learn
• matplotlib

Project Structure

cicddos2019-preprocessing/
├─ notebooks/
│  └─ Preprocessing_Exploitative_Attack_Data_from_the_CICDDOS2019_Dataset.ipynb
├─ docs/
├─ scripts/
├─ .gitignore
├─ LICENSE
├─ README.md
└─ requirements.txt

Roadmap

• Add helper script to preprocess without Jupyter
• Add visualizations for class imbalance
• Add ML baselines on processed data

License

MIT © 2025 Henry Onyeka