Intrusion Detection Systems Using Support Vector Machines on the KDDCUP’99 and NSL-KDD Datasets: A Comprehensive Survey

With the growing rates of cyber-attacks and cyber espionage, need for better more powerful intrusion detection systems (IDS) is even warranted nowadays. The basic task an IDS to act as first line defense, in detecting attacks on internet. As tactics from intruders become sophisticated difficult detect, researchers have started apply novel Machine Learning (ML) techniques effectively detect hence preserve internet users' information overall trust entire network security. Over last decade, there has been explosion research based ML Deep (DL) architectures various security-based datasets such DARPA, KDDCUP'99, NSL-KDD, CAIDA, CTU-13, UNSW-NB15. In this research, we review contemporary literature provide a comprehensive survey different types technique that applies Support Vector Machines (SVMs) algorithms classifier. We focus only studies evaluated two most widely used cybersecurity namely: KDDCUP'99 NSL-KDD datasets. summary each method, identifying role SVMs classifier, all other involved studies. Furthermore, present critical tabular form, highlighting performance measures, strengths, limitations methods surveyed.