This is a project for my thesis for IoT botnet traffic analysis DETECTING, CLASSIFYING AND EXPLAINING IOT BOTNET ATTACKS USING DEEP LEARNING METHODS BASED ON NETWORK DATA
Abstract:
The growing adoption of Internet-of-Things devices brings with it the increased participation of said devices in botnet attacks, and as such novel methods for IoT botnet attack detection are needed. This work demonstrates that deep learning models can be used to detect and classify IoT botnet attacks based on network data in a device agnostic way and that it can be more accurate than some more traditional machine learning methods, especially without feature selection. Furthermore, this works shows that the opaqueness of deep learning models can mitigated to some degree with Local Interpretable Model-Agnostic Explanations technique.
It additionally attempts to reproduce results from this paper https://arxiv.org/abs/1805.03409
This is the dataset used https://archive.ics.uci.edu/ml/machine-learning-databases/00442/