The raw network packets of the UNSW-NB 15 dataset are used in this project to gather information on different type of intrusions. It allows to train several models (SVM, Decision Tree and KNN), that can distinguish intrusion from normal network traffic. The models can be compared using 5-fold cross-validation, and . Some kind of hyperparameter tuning takes place just as well. Also it produces information concerning the features in the dataset. It shows how to find strongly correlated features. It uses a LASSO-technique to find superfluous features. It is more of a starting point for futher study.
-
Notifications
You must be signed in to change notification settings - Fork 0
Using the UNSW-NB15 dataset in practice. This dataset was created to do research on anomaly detection innetwork traffic.
wilfred11/unsw
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
Using the UNSW-NB15 dataset in practice. This dataset was created to do research on anomaly detection innetwork traffic.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published