Using artificial neural network in intrusion detection systems to computer networks

Abstract

The constant growth in the use of computer networks has demanded some concerns regarding disponibility, vulnerability and security. Intrusion Detection Systems (IDS) have been considered essential in keeping network security and therefore have been commonly adopted by network administrators. A possible disadvantage is the fact that such systems are usually based on signature systems, which make them strongly dependent on updated database and consequently inefficient against novel attacks (unknown attacks). The research presented in this paper proposes an IDS system based on artificial neural network (ANN) and the KDDCUP'99 dataset. Experimental results clearly show that the proposed system can reach an overall accuracy of 99.9% regarding the classification of pre-defined classes of intrusion attacks with, which is a very satisfactory result when compared to traditional methods.