Enhancing Cybersecurity with Tailored Datasets: Building an Intrusion Detection Dataset

Abstract

 Datasets play a crucial role in developing and evaluating intrusion detection systems (IDS). These datasets typically contain network traffic data, including both normal and malicious activities, to train and test IDS algorithms. Some widely used datasets for IDS research include KDD Cup 99, NSL-KDD, UNSW-NB15, and CICIDS2017. These datasets provide a diverse range of network attacks, such as denial-of-service (DoS), probe, user-to-root (U2R), and remote-to-local (R2L) attacks. However, it is important to note that many of these datasets are outdated and may not accurately represent current network threats. To address this issue, researchers are continuously developing new datasets that incorporate emerging attack patterns and reflect modern network environments. When selecting a dataset for IDS research, it is crucial to consider factors such as its relevance to current threats, the balance between normal and malicious traffic, and the presence of diverse attack types to ensure the development of robust and effective intrusion detection systems. In this paper different datasets for IDS were discussed along with their characteristics and creation of dataset by simulation of attacks in virtual environment