This project is an implementation of the paper "CapsRule: Explainable Deep Learning for Classifying Network Attacks" https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=10101859. CapsRule is an effective and efficient rule-based deep learning explanation framework dedicated to classifying network attacks. It extracts high-fidelity rules from the feed-forward capsule network that explain how an input sample is classified.
This project is implemented using Python and Tensorflow library.
The repository contains pre-propcessing, training, rule-extraction, rule validation, rule evaluation, and computing performance measure.
This repository contains the source code for the following research article. Please cite this article if you use the code.
Mahdavifar S. and Ghorbani A. A., (2023) CapsRule: Explainable Deep Learning for Classifying Network Attacks”. IEEE Transactions on Neural Networks and Learning Systems. doi=10.1109/TNNLS.2023.3262981: 1-15