Preparing Datasets for Training in a Neural Network System of Intrusion Detection in Industrial Systems

Abstract

One of the commonly known approaches to network security is the intrusion detection system (IDS) that analyzes the behavior of traffic and hosts in the network. This article presents a technique for generating datasets for the IDS and provides their characteristics as well as defines the features of typical attacks against industrial systems. The results can be used for training the AI IDS in detecting security threats to modern industrial systems.