Dataset

This dataset includes a subset of the InSDN Dataset to examine the effects of various flow-to-image conversion techniques on the performance of intrusion detection systems (IDS). The dataset contains five types of attacks: Denial of Service (DoS), Distributed Denial of Service (DDoS), Probe, Normal, and Brute Force Attack (BFA). Each instance represents a network flow, which is converted into an image using: Method 1: applies the Image Generator for Tabular Data (IGTD) framework using Euclidean distance, transforming tabular data into grayscale images. Method 2 extends the IGTD approach but uses Manhattan distance for feature alignment. The dataset also includes a subset of the CIC-DDoS2019 Dataset, which features four types of DDoS attacks: SYN, UDP, MSSQL, and LDAP, alongside BENIGN traffic. For the InSDN dataset, there is one Excel file named InSDN and three folders, each containing images converted by one of the three methods.