Deep Learning based Intrusion Detection on NSL-KDD

The presented model is a neural network solution built with Keras’s Sequential API and contains two experimental models. The model is benchmarked with the NSL-KDD dataset (improved version of the KDD CUP 99 dataset).

Tools

Anaconda (Python 3.6)
Jupyter Notebook

Prerequisites

Keras
Sklearn
Pandas
Numpy

Running the Application

Run inside jupyter notebook after installing required libraries.

Experimental Results

Classification report that contains accuracy per class / attack category available on request.

License

This project is licensed under the MIT License - see the LICENSE.md file for details