ml_classifiers is a Snort 3 Machine Learning-based Inspector for Network Traffic Bi-directional Flow Classification.
It employs several machine learning models previously trained on CICIDS2017 to classify bi-directional flows in real time, completely replacing the Snort 3's default signature-based (or rule-based) detection approach.
Trained classifiers:
- Gaussian/Bernoulli Naive Bayes;
- Linear Support Vector Machine;
- Decision Tree;
- Random Forest;
- AdaBoost.
This project was developed for research purposes of my master's thesis.